mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-07-27 14:38:49 +00:00
0ed41aafbf
# What does this PR do? ## Test Plan pytest tests/verifications/openai_api/test_chat_completion.py --provider openai -k 'test_chat_multiple_images'
3821 lines
254 KiB
JSON
3821 lines
254 KiB
JSON
{
|
|
"created": 1744918192.9299376,
|
|
"duration": 126.91354608535767,
|
|
"exitcode": 1,
|
|
"root": "/home/erichuang/llama-stack",
|
|
"environment": {},
|
|
"summary": {
|
|
"passed": 40,
|
|
"failed": 40,
|
|
"skipped": 4,
|
|
"total": 84,
|
|
"collected": 84
|
|
},
|
|
"collectors": [
|
|
{
|
|
"nodeid": "",
|
|
"outcome": "passed",
|
|
"result": [
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"type": "Module"
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"outcome": "passed",
|
|
"result": [
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-earth]",
|
|
"type": "Function",
|
|
"lineno": 95
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn]",
|
|
"type": "Function",
|
|
"lineno": 95
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]",
|
|
"type": "Function",
|
|
"lineno": 95
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]",
|
|
"type": "Function",
|
|
"lineno": 95
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]",
|
|
"type": "Function",
|
|
"lineno": 95
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]",
|
|
"type": "Function",
|
|
"lineno": 95
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-earth]",
|
|
"type": "Function",
|
|
"lineno": 114
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn]",
|
|
"type": "Function",
|
|
"lineno": 114
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]",
|
|
"type": "Function",
|
|
"lineno": 114
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]",
|
|
"type": "Function",
|
|
"lineno": 114
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]",
|
|
"type": "Function",
|
|
"lineno": 114
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]",
|
|
"type": "Function",
|
|
"lineno": 114
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_image[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 138
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 138
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 138
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_image[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 157
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 157
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 157
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar]",
|
|
"type": "Function",
|
|
"lineno": 181
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-math]",
|
|
"type": "Function",
|
|
"lineno": 181
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]",
|
|
"type": "Function",
|
|
"lineno": 181
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]",
|
|
"type": "Function",
|
|
"lineno": 181
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]",
|
|
"type": "Function",
|
|
"lineno": 181
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]",
|
|
"type": "Function",
|
|
"lineno": 181
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar]",
|
|
"type": "Function",
|
|
"lineno": 204
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-math]",
|
|
"type": "Function",
|
|
"lineno": 204
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]",
|
|
"type": "Function",
|
|
"lineno": 204
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]",
|
|
"type": "Function",
|
|
"lineno": 204
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]",
|
|
"type": "Function",
|
|
"lineno": 204
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]",
|
|
"type": "Function",
|
|
"lineno": 204
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 226
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 226
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 226
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 250
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 250
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 250
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_required[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 278
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 278
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 278
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_required[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 302
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 302
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 302
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 329
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 329
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 329
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"type": "Function",
|
|
"lineno": 352
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"type": "Function",
|
|
"lineno": 352
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"type": "Function",
|
|
"lineno": 352
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]",
|
|
"type": "Function",
|
|
"lineno": 380
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]",
|
|
"type": "Function",
|
|
"lineno": 471
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=False]",
|
|
"type": "Function",
|
|
"lineno": 554
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=True]",
|
|
"type": "Function",
|
|
"lineno": 554
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=False]",
|
|
"type": "Function",
|
|
"lineno": 554
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=True]",
|
|
"type": "Function",
|
|
"lineno": 554
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=False]",
|
|
"type": "Function",
|
|
"lineno": 554
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=True]",
|
|
"type": "Function",
|
|
"lineno": 554
|
|
}
|
|
]
|
|
}
|
|
],
|
|
"tests": [
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-earth]",
|
|
"lineno": 95,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-earth]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-earth",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "earth"
|
|
},
|
|
"setup": {
|
|
"duration": 0.11939296405762434,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6422080835327506,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002934802323579788,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn]",
|
|
"lineno": 95,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "saturn"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07340026367455721,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6134521719068289,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00031049735844135284,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]",
|
|
"lineno": 95,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-earth",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "earth"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07351398840546608,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.898847377859056,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002735760062932968,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]",
|
|
"lineno": 95,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "saturn"
|
|
},
|
|
"setup": {
|
|
"duration": 0.08612977154552937,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6511319326236844,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003559151664376259,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]",
|
|
"lineno": 95,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "earth"
|
|
},
|
|
"setup": {
|
|
"duration": 0.08106738794595003,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.206272155046463,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003584325313568115,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]",
|
|
"lineno": 95,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "saturn"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0796442786231637,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4815350500866771,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00025806669145822525,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-earth]",
|
|
"lineno": 114,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-earth]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-earth",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "earth"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07231954019516706,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.1521263290196657,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00032721273601055145,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn]",
|
|
"lineno": 114,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_basic[meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-saturn",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "saturn"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07364387530833483,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.0600289879366755,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00028987880796194077,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]",
|
|
"lineno": 114,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-earth",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "earth"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07162868417799473,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.2930005770176649,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-earth]>>\nopenai_client = <openai.OpenAI object at 0x7f42743e7760>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'earth', 'input': {'messages': [{'content': 'Which planet do humans live on?', 'role': 'user'}]}, 'output': 'Earth'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_basic\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_basic(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n stream=True,\n )\n content = \"\"\n for chunk in response:\n> content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:132: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0004123607650399208,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]",
|
|
"lineno": 114,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "saturn"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07553945016115904,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4265708066523075,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_basic[meta-llama/Llama-4-Scout-17B-16E-Instruct-saturn]>>\nopenai_client = <openai.OpenAI object at 0x7f42742571f0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'saturn', 'input': {'messages': [{'content': 'Which planet has rings around it with a name starting with letter S?', 'role': 'user'}]}, 'output': 'Saturn'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_basic\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_basic(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n stream=True,\n )\n content = \"\"\n for chunk in response:\n> content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:132: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003767991438508034,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]",
|
|
"lineno": 114,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "earth"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07143466174602509,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.0281891459599137,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-earth]>>\nopenai_client = <openai.OpenAI object at 0x7f4274278310>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'earth', 'input': {'messages': [{'content': 'Which planet do humans live on?', 'role': 'user'}]}, 'output': 'Earth'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_basic\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_basic(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n stream=True,\n )\n content = \"\"\n for chunk in response:\n> content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:132: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003773234784603119,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]",
|
|
"lineno": 114,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "saturn"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07092289440333843,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4124102909117937,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 132,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_basic[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-saturn]>>\nopenai_client = <openai.OpenAI object at 0x7f42743e7310>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'saturn', 'input': {'messages': [{'content': 'Which planet has rings around it with a name starting with letter S?', 'role': 'user'}]}, 'output': 'Saturn'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_basic\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_basic(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n stream=True,\n )\n content = \"\"\n for chunk in response:\n> content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:132: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003204820677638054,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_image[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 138,
|
|
"outcome": "skipped",
|
|
"keywords": [
|
|
"test_chat_non_streaming_image[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07159135863184929,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.0002104705199599266,
|
|
"outcome": "skipped",
|
|
"longrepr": "('/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py', 147, 'Skipped: Skipping test_chat_non_streaming_image for model meta-llama/Llama-3.3-70B-Instruct-Turbo on provider together based on config.')"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003354400396347046,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 138,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0744061404839158,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.2864254424348474,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.000246487557888031,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 138,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07066962588578463,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 4.47614302393049,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00034836214035749435,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_image[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 157,
|
|
"outcome": "skipped",
|
|
"keywords": [
|
|
"test_chat_streaming_image[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.09739464800804853,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.0003191335126757622,
|
|
"outcome": "skipped",
|
|
"longrepr": "('/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py', 166, 'Skipped: Skipping test_chat_streaming_image for model meta-llama/Llama-3.3-70B-Instruct-Turbo on provider together based on config.')"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00026350561529397964,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 157,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.10561292432248592,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.6175378002226353,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 175,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 175,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_image[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f427415f430>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': [{'text': 'What is in this image?', 'type': 'text'}, {'image_url': {...}, 'type': 'image_url'}], 'role': 'user'}]}, 'output': 'llama'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_image\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_image(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n stream=True,\n )\n content = \"\"\n for chunk in response:\n> content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:175: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003682933747768402,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 157,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07195662055164576,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 3.2985631534829736,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 175,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 175,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_image[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42741c7550>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': [{'text': 'What is in this image?', 'type': 'text'}, {'image_url': {...}, 'type': 'image_url'}], 'role': 'user'}]}, 'output': 'llama'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_image\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_image(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n stream=True,\n )\n content = \"\"\n for chunk in response:\n> content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:175: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003777453675866127,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar]",
|
|
"lineno": 181,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "calendar"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0733196372166276,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.40959454514086246,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00029125437140464783,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-math]",
|
|
"lineno": 181,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-math]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-math",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "math"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07248916011303663,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 3.498455540277064,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00023921672254800797,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]",
|
|
"lineno": 181,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "calendar"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07911352813243866,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6717434097081423,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00025916099548339844,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]",
|
|
"lineno": 181,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-math",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "math"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07156322989612818,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 3.698870756663382,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002654632553458214,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]",
|
|
"lineno": 181,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "calendar"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07457748707383871,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.8891718471422791,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002395138144493103,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]",
|
|
"lineno": 181,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "math"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07155069429427385,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 3.276700599119067,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002568913623690605,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar]",
|
|
"lineno": 204,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-calendar",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "calendar"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07365360390394926,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.7638470390811563,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00027653202414512634,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-math]",
|
|
"lineno": 204,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_structured_output[meta-llama/Llama-3.3-70B-Instruct-Turbo-math]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-math",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "math"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07424602191895247,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 3.622116087935865,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002861013635993004,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]",
|
|
"lineno": 204,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "calendar"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07192372716963291,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.5049019353464246,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-calendar]>>\nopenai_client = <openai.OpenAI object at 0x7f4274178c10>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'calendar', 'input': {'messages': [{'content': 'Extract the event information.', 'role': 'system'}, {'cont...articipants'], 'title': 'CalendarEvent', 'type': 'object'}}, 'type': 'json_schema'}}, 'output': 'valid_calendar_event'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_structured_output\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_structured_output(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n response_format=case[\"input\"][\"response_format\"],\n stream=True,\n )\n maybe_json_content = \"\"\n for chunk in response:\n> maybe_json_content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:223: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00036794692277908325,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]",
|
|
"lineno": 204,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-math",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "math"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07304532174021006,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.961389934644103,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_structured_output[meta-llama/Llama-4-Scout-17B-16E-Instruct-math]>>\nopenai_client = <openai.OpenAI object at 0x7f42741786d0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'math', 'input': {'messages': [{'content': 'You are a helpful math tutor. Guide the user through the solut... ['steps', 'final_answer'], 'title': 'MathReasoning', ...}}, 'type': 'json_schema'}}, 'output': 'valid_math_reasoning'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_structured_output\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_structured_output(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n response_format=case[\"input\"][\"response_format\"],\n stream=True,\n )\n maybe_json_content = \"\"\n for chunk in response:\n> maybe_json_content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:223: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003312695771455765,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]",
|
|
"lineno": 204,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "calendar"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07350922282785177,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6764275450259447,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-calendar]>>\nopenai_client = <openai.OpenAI object at 0x7f427420ff40>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'calendar', 'input': {'messages': [{'content': 'Extract the event information.', 'role': 'system'}, {'cont...articipants'], 'title': 'CalendarEvent', 'type': 'object'}}, 'type': 'json_schema'}}, 'output': 'valid_calendar_event'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_structured_output\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_structured_output(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n response_format=case[\"input\"][\"response_format\"],\n stream=True,\n )\n maybe_json_content = \"\"\n for chunk in response:\n> maybe_json_content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:223: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003826189786195755,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]",
|
|
"lineno": 204,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "math"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07295230869203806,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 10.689278944395483,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 223,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_structured_output[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-math]>>\nopenai_client = <openai.OpenAI object at 0x7f427415eb60>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'math', 'input': {'messages': [{'content': 'You are a helpful math tutor. Guide the user through the solut... ['steps', 'final_answer'], 'title': 'MathReasoning', ...}}, 'type': 'json_schema'}}, 'output': 'valid_math_reasoning'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_chat_structured_output\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_structured_output(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n response_format=case[\"input\"][\"response_format\"],\n stream=True,\n )\n maybe_json_content = \"\"\n for chunk in response:\n> maybe_json_content += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:223: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0004014279693365097,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 226,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.09202722646296024,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.8140280386433005,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003595082089304924,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 226,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.09484888892620802,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.3706049248576164,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003290809690952301,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 226,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.10521113499999046,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.36842701490968466,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00031410157680511475,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 250,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.10422383341938257,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6454980997368693,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002997415140271187,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 250,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.09408890828490257,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.36066764686256647,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 268,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42741c44f0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_calling(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n stream=True,\n )\n \n> _, tool_calls_buffer = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:268: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f4274268760>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00035039614886045456,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 250,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07232134602963924,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4706049496307969,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 268,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f427417ee60>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"],\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_calling(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n stream=True,\n )\n \n> _, tool_calls_buffer = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:268: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f427416d960>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00039384420961141586,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_required[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 278,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_choice_required[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07465469185262918,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4374591317027807,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003099888563156128,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 278,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07351493183523417,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4368853671476245,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00026369933038949966,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 278,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07258845027536154,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.940508272498846,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00032961275428533554,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_required[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 302,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_choice_required[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07273276895284653,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6150273764505982,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002876110374927521,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 302,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07505382597446442,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.5026597818359733,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 321,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42742aa050>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_choice_required(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"required\", # Force tool call\n stream=True,\n )\n \n> _, tool_calls_buffer = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:321: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f42741e9810>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003487151116132736,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 302,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07343385275453329,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.720921658910811,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 321,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_choice_required[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f427416dab0>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_choice_required(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"required\", # Force tool call\n stream=True,\n )\n \n> _, tool_calls_buffer = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:321: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f427447c340>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0004109758883714676,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 329,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07189673464745283,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.403152690269053,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 349,
|
|
"message": "AssertionError: Expected no tool calls when tool_choice='none'\nassert [ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)] is None\n + where [ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)]).tool_calls\n + where ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)]) = Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)]), seed=4867562177231181000).message"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 349,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42741eb670>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_non_streaming_tool_choice_none(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"none\",\n stream=False,\n )\n \n assert response.choices[0].message.role == \"assistant\"\n> assert response.choices[0].message.tool_calls is None, \"Expected no tool calls when tool_choice='none'\"\nE AssertionError: Expected no tool calls when tool_choice='none'\nE assert [ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)] is None\nE + where [ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)]).tool_calls\nE + where ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)]) = Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_xx4eg2o4wladhs7i0gy8d2cb', function=Function(arguments='{\"location\":\"San Francisco, USA\"}', name='get_weather'), type='function', index=0)]), seed=4867562177231181000).message\n\ntests/verifications/openai_api/test_chat_completion.py:349: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00037758704274892807,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 329,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07282305508852005,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4538485202938318,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 349,
|
|
"message": "AssertionError: Expected no tool calls when tool_choice='none'\nassert [ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] is None\n + where [ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]).tool_calls\n + where ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]) = Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]), seed=None).message"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 349,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f4274247160>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_non_streaming_tool_choice_none(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"none\",\n stream=False,\n )\n \n assert response.choices[0].message.role == \"assistant\"\n> assert response.choices[0].message.tool_calls is None, \"Expected no tool calls when tool_choice='none'\"\nE AssertionError: Expected no tool calls when tool_choice='none'\nE assert [ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] is None\nE + where [ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]).tool_calls\nE + where ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]) = Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_6gehr7flf4gaqu65prmi1pca', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]), seed=None).message\n\ntests/verifications/openai_api/test_chat_completion.py:349: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003799665719270706,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 329,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07050042506307364,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.3740060832351446,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 349,
|
|
"message": "AssertionError: Expected no tool calls when tool_choice='none'\nassert [ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] is None\n + where [ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]).tool_calls\n + where ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]) = Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]), seed=None).message"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 349,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42742f3220>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_non_streaming_tool_choice_none(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n response = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"none\",\n stream=False,\n )\n \n assert response.choices[0].message.role == \"assistant\"\n> assert response.choices[0].message.tool_calls is None, \"Expected no tool calls when tool_choice='none'\"\nE AssertionError: Expected no tool calls when tool_choice='none'\nE assert [ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] is None\nE + where [ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]).tool_calls\nE + where ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]) = Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_ngwnt1xmgxipkswdhdepisni', function=Function(arguments='{\"location\":\"San Francisco\"}', name='get_weather'), type='function', index=0)]), seed=None).message\n\ntests/verifications/openai_api/test_chat_completion.py:349: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003066370263695717,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"lineno": 352,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.06983672920614481,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6774894064292312,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 376,
|
|
"message": "AssertionError: Expected no tool call chunks when tool_choice='none'\nassert not [ChoiceDeltaToolCall(index=0, id='call_emdpbpvm77rqbzz66arrzv5w', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]\n + where [ChoiceDeltaToolCall(index=0, id='call_emdpbpvm77rqbzz66arrzv5w', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')] = ChoiceDelta(content=None, function_call=None, refusal=None, role=None, tool_calls=[ChoiceDeltaToolCall(index=0, id='call_emdpbpvm77rqbzz66arrzv5w', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]).tool_calls"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 376,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_choice_none[meta-llama/Llama-3.3-70B-Instruct-Turbo-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f427430d480>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_choice_none(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"none\",\n stream=True,\n )\n \n content = \"\"\n for chunk in stream:\n delta = chunk.choices[0].delta\n if delta.content:\n content += delta.content\n> assert not delta.tool_calls, \"Expected no tool call chunks when tool_choice='none'\"\nE AssertionError: Expected no tool call chunks when tool_choice='none'\nE assert not [ChoiceDeltaToolCall(index=0, id='call_emdpbpvm77rqbzz66arrzv5w', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]\nE + where [ChoiceDeltaToolCall(index=0, id='call_emdpbpvm77rqbzz66arrzv5w', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')] = ChoiceDelta(content=None, function_call=None, refusal=None, role=None, tool_calls=[ChoiceDeltaToolCall(index=0, id='call_emdpbpvm77rqbzz66arrzv5w', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]).tool_calls\n\ntests/verifications/openai_api/test_chat_completion.py:376: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003580348566174507,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"lineno": 352,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07331710867583752,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.38044120091944933,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 376,
|
|
"message": "AssertionError: Expected no tool call chunks when tool_choice='none'\nassert not [ChoiceDeltaToolCall(index=0, id='call_g85q6ysacljgjczgq8r30tjv', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]\n + where [ChoiceDeltaToolCall(index=0, id='call_g85q6ysacljgjczgq8r30tjv', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')] = ChoiceDelta(content=None, function_call=None, refusal=None, role=None, tool_calls=[ChoiceDeltaToolCall(index=0, id='call_g85q6ysacljgjczgq8r30tjv', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]).tool_calls"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 376,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Scout-17B-16E-Instruct-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42745f3970>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_choice_none(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"none\",\n stream=True,\n )\n \n content = \"\"\n for chunk in stream:\n delta = chunk.choices[0].delta\n if delta.content:\n content += delta.content\n> assert not delta.tool_calls, \"Expected no tool call chunks when tool_choice='none'\"\nE AssertionError: Expected no tool call chunks when tool_choice='none'\nE assert not [ChoiceDeltaToolCall(index=0, id='call_g85q6ysacljgjczgq8r30tjv', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]\nE + where [ChoiceDeltaToolCall(index=0, id='call_g85q6ysacljgjczgq8r30tjv', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')] = ChoiceDelta(content=None, function_call=None, refusal=None, role=None, tool_calls=[ChoiceDeltaToolCall(index=0, id='call_g85q6ysacljgjczgq8r30tjv', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]).tool_calls\n\ntests/verifications/openai_api/test_chat_completion.py:376: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003765234723687172,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"lineno": 352,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "case0"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07194581907242537,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.37374384608119726,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 376,
|
|
"message": "AssertionError: Expected no tool call chunks when tool_choice='none'\nassert not [ChoiceDeltaToolCall(index=0, id='call_zq6x10vfu9pkxme6pm9zxouk', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]\n + where [ChoiceDeltaToolCall(index=0, id='call_zq6x10vfu9pkxme6pm9zxouk', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')] = ChoiceDelta(content=None, function_call=None, refusal=None, role=None, tool_calls=[ChoiceDeltaToolCall(index=0, id='call_zq6x10vfu9pkxme6pm9zxouk', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]).tool_calls"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 376,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_tool_choice_none[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-case0]>>\nopenai_client = <openai.OpenAI object at 0x7f42741c4520>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'input': {'messages': [{'content': 'You are a helpful assistant that can use tools to get information.', 'role': 'sys..., 'properties': {...}, 'required': [...], 'type': 'object'}}, 'type': 'function'}]}, 'output': 'get_weather_tool_call'}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases[\"test_tool_calling\"][\"test_params\"][\"case\"], # Reusing existing case for now\n ids=case_id_generator,\n )\n def test_chat_streaming_tool_choice_none(request, openai_client, model, provider, verification_config, case):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n stream = openai_client.chat.completions.create(\n model=model,\n messages=case[\"input\"][\"messages\"],\n tools=case[\"input\"][\"tools\"],\n tool_choice=\"none\",\n stream=True,\n )\n \n content = \"\"\n for chunk in stream:\n delta = chunk.choices[0].delta\n if delta.content:\n content += delta.content\n> assert not delta.tool_calls, \"Expected no tool call chunks when tool_choice='none'\"\nE AssertionError: Expected no tool call chunks when tool_choice='none'\nE assert not [ChoiceDeltaToolCall(index=0, id='call_zq6x10vfu9pkxme6pm9zxouk', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]\nE + where [ChoiceDeltaToolCall(index=0, id='call_zq6x10vfu9pkxme6pm9zxouk', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')] = ChoiceDelta(content=None, function_call=None, refusal=None, role=None, tool_calls=[ChoiceDeltaToolCall(index=0, id='call_zq6x10vfu9pkxme6pm9zxouk', function=ChoiceDeltaToolCallFunction(arguments='', name='get_weather'), type='function')]).tool_calls\n\ntests/verifications/openai_api/test_chat_completion.py:376: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003813542425632477,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]",
|
|
"lineno": 380,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "text_then_weather_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07330320309847593,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.4314677305519581,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 439,
|
|
"message": "AssertionError: Expected 0 tool calls, but got 1\nassert 1 == 0\n + where 1 = len(([ChatCompletionMessageToolCall(id='call_l05cckdk5mooai2iyfucg4s8', function=Function(arguments='{\"location\":\"San Francisco, CA\"}', name='get_weather'), type='function', index=0)]))\n + where [ChatCompletionMessageToolCall(id='call_l05cckdk5mooai2iyfucg4s8', function=Function(arguments='{\"location\":\"San Francisco, CA\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_l05cckdk5mooai2iyfucg4s8', function=Function(arguments='{\"location\":\"San Francisco, CA\"}', name='get_weather'), type='function', index=0)]).tool_calls"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 439,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f4274148ca0>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'text_then_weather_tool', 'expected': [{'answer': ['sol'], 'num_tool_calls': 0}, {'num_tool_calls': 1, 'to...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_non_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\"\n Test cases for multi-turn tool calling.\n Tool calls are asserted.\n Tool responses are provided in the test case.\n Final response is asserted.\n \"\"\"\n \n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n # Create a copy of the messages list to avoid modifying the original\n messages = []\n tools = case[\"input\"][\"tools\"]\n # Use deepcopy to prevent modification across runs/parametrization\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n # keep going until either\n # 1. we have messages to test in multi-turn\n # 2. no messages but last message is tool response\n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n # do not take new messages if last message is tool response\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n # Ensure new_messages is a list of message objects\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n # If it's a single message object, add it directly\n messages.append(new_messages)\n \n # --- API Call ---\n response = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=False,\n )\n \n # --- Process Response ---\n assistant_message = response.choices[0].message\n messages.append(assistant_message.model_dump(exclude_unset=True))\n \n assert assistant_message.role == \"assistant\"\n \n # Get the expected result data\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n # --- Assertions based on expected result ---\n> assert len(assistant_message.tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(assistant_message.tool_calls or [])}\"\n )\nE AssertionError: Expected 0 tool calls, but got 1\nE assert 1 == 0\nE + where 1 = len(([ChatCompletionMessageToolCall(id='call_l05cckdk5mooai2iyfucg4s8', function=Function(arguments='{\"location\":\"San Francisco, CA\"}', name='get_weather'), type='function', index=0)]))\nE + where [ChatCompletionMessageToolCall(id='call_l05cckdk5mooai2iyfucg4s8', function=Function(arguments='{\"location\":\"San Francisco, CA\"}', name='get_weather'), type='function', index=0)] = ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_l05cckdk5mooai2iyfucg4s8', function=Function(arguments='{\"location\":\"San Francisco, CA\"}', name='get_weather'), type='function', index=0)]).tool_calls\n\ntests/verifications/openai_api/test_chat_completion.py:439: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00040314625948667526,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "weather_tool_then_text"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07405277714133263,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.8350177155807614,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00023361947387456894,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "add_product_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07361320778727531,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.0619212854653597,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002395985648036003,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "get_then_create_event_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07290417980402708,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 4.241749887354672,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00027841050177812576,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "compare_monthly_expense_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07301546633243561,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.0520667918026447,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002469858154654503,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]",
|
|
"lineno": 380,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "text_then_weather_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07405530381947756,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.48041669093072414,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 467,
|
|
"message": "AssertionError: Expected one of ['sol'] in content, but got: 'I am not able to complete this task as it falls outside of the scope of the functions I have been given.'\nassert False\n + where False = any(<generator object test_chat_non_streaming_multi_turn_tool_calling.<locals>.<genexpr> at 0x7f4274057610>)"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 467,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f42740f7700>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'text_then_weather_tool', 'expected': [{'answer': ['sol'], 'num_tool_calls': 0}, {'num_tool_calls': 1, 'to...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_non_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\"\n Test cases for multi-turn tool calling.\n Tool calls are asserted.\n Tool responses are provided in the test case.\n Final response is asserted.\n \"\"\"\n \n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n # Create a copy of the messages list to avoid modifying the original\n messages = []\n tools = case[\"input\"][\"tools\"]\n # Use deepcopy to prevent modification across runs/parametrization\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n # keep going until either\n # 1. we have messages to test in multi-turn\n # 2. no messages but last message is tool response\n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n # do not take new messages if last message is tool response\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n # Ensure new_messages is a list of message objects\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n # If it's a single message object, add it directly\n messages.append(new_messages)\n \n # --- API Call ---\n response = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=False,\n )\n \n # --- Process Response ---\n assistant_message = response.choices[0].message\n messages.append(assistant_message.model_dump(exclude_unset=True))\n \n assert assistant_message.role == \"assistant\"\n \n # Get the expected result data\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n # --- Assertions based on expected result ---\n assert len(assistant_message.tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(assistant_message.tool_calls or [])}\"\n )\n \n if num_tool_calls > 0:\n tool_call = assistant_message.tool_calls[0]\n assert tool_call.function.name == expected[\"tool_name\"], (\n f\"Expected tool '{expected['tool_name']}', got '{tool_call.function.name}'\"\n )\n # Parse the JSON string arguments before comparing\n actual_arguments = json.loads(tool_call.function.arguments)\n assert actual_arguments == expected[\"tool_arguments\"], (\n f\"Expected arguments '{expected['tool_arguments']}', got '{actual_arguments}'\"\n )\n \n # Prepare and append the tool response for the next turn\n tool_response = tool_responses.pop(0)\n messages.append(\n {\n \"role\": \"tool\",\n \"tool_call_id\": tool_call.id,\n \"content\": tool_response[\"response\"],\n }\n )\n else:\n assert assistant_message.content is not None, \"Expected content, but none received.\"\n expected_answers = expected[\"answer\"] # This is now a list\n content_lower = assistant_message.content.lower()\n> assert any(ans.lower() in content_lower for ans in expected_answers), (\n f\"Expected one of {expected_answers} in content, but got: '{assistant_message.content}'\"\n )\nE AssertionError: Expected one of ['sol'] in content, but got: 'I am not able to complete this task as it falls outside of the scope of the functions I have been given.'\nE assert False\nE + where False = any(<generator object test_chat_non_streaming_multi_turn_tool_calling.<locals>.<genexpr> at 0x7f4274057610>)\n\ntests/verifications/openai_api/test_chat_completion.py:467: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00035319291055202484,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "weather_tool_then_text"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0724497502669692,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.832760401070118,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00026283878833055496,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "add_product_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07180811651051044,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.4359142612665892,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002761436626315117,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "get_then_create_event_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07503274269402027,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.909641013480723,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002613905817270279,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "compare_monthly_expense_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07153380755335093,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.695867782458663,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00032124295830726624,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]",
|
|
"lineno": 380,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "text_then_weather_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07275318540632725,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.34551760647445917,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 467,
|
|
"message": "AssertionError: Expected one of ['sol'] in content, but got: '{\"name\": null, \"parameters\": null}'\nassert False\n + where False = any(<generator object test_chat_non_streaming_multi_turn_tool_calling.<locals>.<genexpr> at 0x7f42742dd4d0>)"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 467,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f427414b970>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'text_then_weather_tool', 'expected': [{'answer': ['sol'], 'num_tool_calls': 0}, {'num_tool_calls': 1, 'to...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_non_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\"\n Test cases for multi-turn tool calling.\n Tool calls are asserted.\n Tool responses are provided in the test case.\n Final response is asserted.\n \"\"\"\n \n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n # Create a copy of the messages list to avoid modifying the original\n messages = []\n tools = case[\"input\"][\"tools\"]\n # Use deepcopy to prevent modification across runs/parametrization\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n # keep going until either\n # 1. we have messages to test in multi-turn\n # 2. no messages but last message is tool response\n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n # do not take new messages if last message is tool response\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n # Ensure new_messages is a list of message objects\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n # If it's a single message object, add it directly\n messages.append(new_messages)\n \n # --- API Call ---\n response = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=False,\n )\n \n # --- Process Response ---\n assistant_message = response.choices[0].message\n messages.append(assistant_message.model_dump(exclude_unset=True))\n \n assert assistant_message.role == \"assistant\"\n \n # Get the expected result data\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n # --- Assertions based on expected result ---\n assert len(assistant_message.tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(assistant_message.tool_calls or [])}\"\n )\n \n if num_tool_calls > 0:\n tool_call = assistant_message.tool_calls[0]\n assert tool_call.function.name == expected[\"tool_name\"], (\n f\"Expected tool '{expected['tool_name']}', got '{tool_call.function.name}'\"\n )\n # Parse the JSON string arguments before comparing\n actual_arguments = json.loads(tool_call.function.arguments)\n assert actual_arguments == expected[\"tool_arguments\"], (\n f\"Expected arguments '{expected['tool_arguments']}', got '{actual_arguments}'\"\n )\n \n # Prepare and append the tool response for the next turn\n tool_response = tool_responses.pop(0)\n messages.append(\n {\n \"role\": \"tool\",\n \"tool_call_id\": tool_call.id,\n \"content\": tool_response[\"response\"],\n }\n )\n else:\n assert assistant_message.content is not None, \"Expected content, but none received.\"\n expected_answers = expected[\"answer\"] # This is now a list\n content_lower = assistant_message.content.lower()\n> assert any(ans.lower() in content_lower for ans in expected_answers), (\n f\"Expected one of {expected_answers} in content, but got: '{assistant_message.content}'\"\n )\nE AssertionError: Expected one of ['sol'] in content, but got: '{\"name\": null, \"parameters\": null}'\nE assert False\nE + where False = any(<generator object test_chat_non_streaming_multi_turn_tool_calling.<locals>.<genexpr> at 0x7f42742dd4d0>)\n\ntests/verifications/openai_api/test_chat_completion.py:467: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003842068836092949,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "weather_tool_then_text"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07281951513141394,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.008104412816465,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00026233773678541183,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "add_product_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07155719958245754,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.3485742239281535,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002629430964589119,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]",
|
|
"lineno": 380,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "get_then_create_event_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07251190021634102,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.9882029946893454,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 450,
|
|
"message": "AssertionError: Expected arguments '{'name': 'Team Building', 'date': '2025-03-03', 'time': '10:00', 'location': 'Main Conference Room', 'participants': ['Alice', 'Bob', 'Charlie']}', got '{'date': '\"2025-03-03\"', 'location': '\"Main Conference Room\"', 'name': '\"Team Building\"', 'participants': ['Alice', 'Bob', 'Charlie'], 'time': '\"10:00\"'}'\nassert {'date': '\"20...harlie'], ...} == {'date': '202...harlie'], ...}\n \n Omitting 1 identical items, use -vv to show\n Differing items:\n {'date': '\"2025-03-03\"'} != {'date': '2025-03-03'}\n {'name': '\"Team Building\"'} != {'name': 'Team Building'}\n {'time': '\"10:00\"'} != {'time': '10:00'}\n {'location': '\"Main Conference Room\"'} != {'location': 'Main Conference Room'}...\n \n ...Full output truncated (21 lines hidden), use '-vv' to show"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 450,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f4274027af0>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'get_then_create_event_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'date': '2025-03-03', ...ents found for 2025-03-03 at 10:00'}\"}, {'response': \"{'response': 'Successfully created new event with id: e_123'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_non_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\"\n Test cases for multi-turn tool calling.\n Tool calls are asserted.\n Tool responses are provided in the test case.\n Final response is asserted.\n \"\"\"\n \n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n # Create a copy of the messages list to avoid modifying the original\n messages = []\n tools = case[\"input\"][\"tools\"]\n # Use deepcopy to prevent modification across runs/parametrization\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n # keep going until either\n # 1. we have messages to test in multi-turn\n # 2. no messages but last message is tool response\n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n # do not take new messages if last message is tool response\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n # Ensure new_messages is a list of message objects\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n # If it's a single message object, add it directly\n messages.append(new_messages)\n \n # --- API Call ---\n response = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=False,\n )\n \n # --- Process Response ---\n assistant_message = response.choices[0].message\n messages.append(assistant_message.model_dump(exclude_unset=True))\n \n assert assistant_message.role == \"assistant\"\n \n # Get the expected result data\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n # --- Assertions based on expected result ---\n assert len(assistant_message.tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(assistant_message.tool_calls or [])}\"\n )\n \n if num_tool_calls > 0:\n tool_call = assistant_message.tool_calls[0]\n assert tool_call.function.name == expected[\"tool_name\"], (\n f\"Expected tool '{expected['tool_name']}', got '{tool_call.function.name}'\"\n )\n # Parse the JSON string arguments before comparing\n actual_arguments = json.loads(tool_call.function.arguments)\n> assert actual_arguments == expected[\"tool_arguments\"], (\n f\"Expected arguments '{expected['tool_arguments']}', got '{actual_arguments}'\"\n )\nE AssertionError: Expected arguments '{'name': 'Team Building', 'date': '2025-03-03', 'time': '10:00', 'location': 'Main Conference Room', 'participants': ['Alice', 'Bob', 'Charlie']}', got '{'date': '\"2025-03-03\"', 'location': '\"Main Conference Room\"', 'name': '\"Team Building\"', 'participants': ['Alice', 'Bob', 'Charlie'], 'time': '\"10:00\"'}'\nE assert {'date': '\"20...harlie'], ...} == {'date': '202...harlie'], ...}\nE \nE Omitting 1 identical items, use -vv to show\nE Differing items:\nE {'date': '\"2025-03-03\"'} != {'date': '2025-03-03'}\nE {'name': '\"Team Building\"'} != {'name': 'Team Building'}\nE {'time': '\"10:00\"'} != {'time': '10:00'}\nE {'location': '\"Main Conference Room\"'} != {'location': 'Main Conference Room'}...\nE \nE ...Full output truncated (21 lines hidden), use '-vv' to show\n\ntests/verifications/openai_api/test_chat_completion.py:450: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003328891471028328,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]",
|
|
"lineno": 380,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_non_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "compare_monthly_expense_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07363704219460487,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 4.031332626007497,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002817586064338684,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "text_then_weather_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07673048228025436,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.3994998000562191,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 521,
|
|
"message": "AssertionError: Expected 0 tool calls, but got 1\nassert 1 == 0\n + where 1 = len(([{'function': {'arguments': '{\"location\":\"San Francisco, CA\"}', 'name': 'get_weather'}, 'id': 'call_dqcu28a6iyxlobv36c23k0qp', 'type': 'function'}]))"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 521,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-text_then_weather_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f4274179c30>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'text_then_weather_tool', 'expected': [{'answer': ['sol'], 'num_tool_calls': 0}, {'num_tool_calls': 1, 'to...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n \n # --- Construct Assistant Message for History ---\n assistant_message_dict = {\"role\": \"assistant\"}\n if accumulated_content:\n assistant_message_dict[\"content\"] = accumulated_content\n if accumulated_tool_calls:\n assistant_message_dict[\"tool_calls\"] = accumulated_tool_calls\n \n messages.append(assistant_message_dict)\n \n # --- Assertions ---\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n> assert len(accumulated_tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(accumulated_tool_calls or [])}\"\n )\nE AssertionError: Expected 0 tool calls, but got 1\nE assert 1 == 0\nE + where 1 = len(([{'function': {'arguments': '{\"location\":\"San Francisco, CA\"}', 'name': 'get_weather'}, 'id': 'call_dqcu28a6iyxlobv36c23k0qp', 'type': 'function'}]))\n\ntests/verifications/openai_api/test_chat_completion.py:521: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003687366843223572,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "weather_tool_then_text"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07477510999888182,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.918418399989605,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 547,
|
|
"message": "AssertionError: Expected content, but none received.\nassert ('' is not None and '' != '')"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 547,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-weather_tool_then_text]>>\nopenai_client = <openai.OpenAI object at 0x7f427417a2c0>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'weather_tool_then_text', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'location': 'San Francisco...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n \n # --- Construct Assistant Message for History ---\n assistant_message_dict = {\"role\": \"assistant\"}\n if accumulated_content:\n assistant_message_dict[\"content\"] = accumulated_content\n if accumulated_tool_calls:\n assistant_message_dict[\"tool_calls\"] = accumulated_tool_calls\n \n messages.append(assistant_message_dict)\n \n # --- Assertions ---\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n assert len(accumulated_tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(accumulated_tool_calls or [])}\"\n )\n \n if num_tool_calls > 0:\n # Use the first accumulated tool call for assertion\n tool_call = accumulated_tool_calls[0]\n assert tool_call[\"function\"][\"name\"] == expected[\"tool_name\"], (\n f\"Expected tool '{expected['tool_name']}', got '{tool_call['function']['name']}'\"\n )\n # Parse the accumulated arguments string for comparison\n actual_arguments = json.loads(tool_call[\"function\"][\"arguments\"])\n assert actual_arguments == expected[\"tool_arguments\"], (\n f\"Expected arguments '{expected['tool_arguments']}', got '{actual_arguments}'\"\n )\n \n # Prepare and append the tool response for the next turn\n tool_response = tool_responses.pop(0)\n messages.append(\n {\n \"role\": \"tool\",\n \"tool_call_id\": tool_call[\"id\"],\n \"content\": tool_response[\"response\"],\n }\n )\n else:\n> assert accumulated_content is not None and accumulated_content != \"\", \"Expected content, but none received.\"\nE AssertionError: Expected content, but none received.\nE assert ('' is not None and '' != '')\n\ntests/verifications/openai_api/test_chat_completion.py:547: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00036141276359558105,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool]",
|
|
"lineno": 471,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-add_product_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "add_product_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07217607088387012,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.2676455974578857,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00024215038865804672,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "get_then_create_event_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0713065592572093,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.0453352769836783,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 547,
|
|
"message": "AssertionError: Expected content, but none received.\nassert ('' is not None and '' != '')"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 547,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-get_then_create_event_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f427415e0b0>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'get_then_create_event_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'date': '2025-03-03', ...ents found for 2025-03-03 at 10:00'}\"}, {'response': \"{'response': 'Successfully created new event with id: e_123'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n \n # --- Construct Assistant Message for History ---\n assistant_message_dict = {\"role\": \"assistant\"}\n if accumulated_content:\n assistant_message_dict[\"content\"] = accumulated_content\n if accumulated_tool_calls:\n assistant_message_dict[\"tool_calls\"] = accumulated_tool_calls\n \n messages.append(assistant_message_dict)\n \n # --- Assertions ---\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n assert len(accumulated_tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(accumulated_tool_calls or [])}\"\n )\n \n if num_tool_calls > 0:\n # Use the first accumulated tool call for assertion\n tool_call = accumulated_tool_calls[0]\n assert tool_call[\"function\"][\"name\"] == expected[\"tool_name\"], (\n f\"Expected tool '{expected['tool_name']}', got '{tool_call['function']['name']}'\"\n )\n # Parse the accumulated arguments string for comparison\n actual_arguments = json.loads(tool_call[\"function\"][\"arguments\"])\n assert actual_arguments == expected[\"tool_arguments\"], (\n f\"Expected arguments '{expected['tool_arguments']}', got '{actual_arguments}'\"\n )\n \n # Prepare and append the tool response for the next turn\n tool_response = tool_responses.pop(0)\n messages.append(\n {\n \"role\": \"tool\",\n \"tool_call_id\": tool_call[\"id\"],\n \"content\": tool_response[\"response\"],\n }\n )\n else:\n> assert accumulated_content is not None and accumulated_content != \"\", \"Expected content, but none received.\"\nE AssertionError: Expected content, but none received.\nE assert ('' is not None and '' != '')\n\ntests/verifications/openai_api/test_chat_completion.py:547: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00030668359249830246,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "compare_monthly_expense_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07108221855014563,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.034472893923521,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 547,
|
|
"message": "AssertionError: Expected content, but none received.\nassert ('' is not None and '' != '')"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 547,
|
|
"message": "AssertionError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-3.3-70B-Instruct-Turbo-compare_monthly_expense_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f42743b7a90>\nmodel = 'meta-llama/Llama-3.3-70B-Instruct-Turbo', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'compare_monthly_expense_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'month': 1, 'year': ... 'Total expenses for January 2025: $1000'}\"}, {'response': \"{'response': 'Total expenses for February 2024: $2000'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n \n # --- Construct Assistant Message for History ---\n assistant_message_dict = {\"role\": \"assistant\"}\n if accumulated_content:\n assistant_message_dict[\"content\"] = accumulated_content\n if accumulated_tool_calls:\n assistant_message_dict[\"tool_calls\"] = accumulated_tool_calls\n \n messages.append(assistant_message_dict)\n \n # --- Assertions ---\n expected = expected_results.pop(0)\n num_tool_calls = expected[\"num_tool_calls\"]\n \n assert len(accumulated_tool_calls or []) == num_tool_calls, (\n f\"Expected {num_tool_calls} tool calls, but got {len(accumulated_tool_calls or [])}\"\n )\n \n if num_tool_calls > 0:\n # Use the first accumulated tool call for assertion\n tool_call = accumulated_tool_calls[0]\n assert tool_call[\"function\"][\"name\"] == expected[\"tool_name\"], (\n f\"Expected tool '{expected['tool_name']}', got '{tool_call['function']['name']}'\"\n )\n # Parse the accumulated arguments string for comparison\n actual_arguments = json.loads(tool_call[\"function\"][\"arguments\"])\n assert actual_arguments == expected[\"tool_arguments\"], (\n f\"Expected arguments '{expected['tool_arguments']}', got '{actual_arguments}'\"\n )\n \n # Prepare and append the tool response for the next turn\n tool_response = tool_responses.pop(0)\n messages.append(\n {\n \"role\": \"tool\",\n \"tool_call_id\": tool_call[\"id\"],\n \"content\": tool_response[\"response\"],\n }\n )\n else:\n> assert accumulated_content is not None and accumulated_content != \"\", \"Expected content, but none received.\"\nE AssertionError: Expected content, but none received.\nE assert ('' is not None and '' != '')\n\ntests/verifications/openai_api/test_chat_completion.py:547: AssertionError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00035398639738559723,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "text_then_weather_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07186305243521929,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.8766405330970883,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-text_then_weather_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f42743e54b0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'text_then_weather_tool', 'expected': [{'answer': ['sol'], 'num_tool_calls': 0}, {'num_tool_calls': 1, 'to...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f42742f0820>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003088880330324173,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "weather_tool_then_text"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0846314700320363,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.40889575984328985,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-weather_tool_then_text]>>\nopenai_client = <openai.OpenAI object at 0x7f42742f2bc0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'weather_tool_then_text', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'location': 'San Francisco...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f42740fd270>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003652172163128853,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "add_product_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07273881137371063,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.251293654553592,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-add_product_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f427420eda0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'add_product_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'inStock': True, 'name': 'Widget...}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': 'Successfully added product with id: 123'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f4273f940a0>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00030664633959531784,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "get_then_create_event_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.071181770414114,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.5708655547350645,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-get_then_create_event_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f42740fc910>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'get_then_create_event_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'date': '2025-03-03', ...ents found for 2025-03-03 at 10:00'}\"}, {'response': \"{'response': 'Successfully created new event with id: e_123'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f4273f82b90>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00036500580608844757,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "compare_monthly_expense_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.06934114638715982,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.5055103581398726,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Scout-17B-16E-Instruct-compare_monthly_expense_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f427410dea0>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'compare_monthly_expense_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'month': 1, 'year': ... 'Total expenses for January 2025: $1000'}\"}, {'response': \"{'response': 'Total expenses for February 2024: $2000'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f427430c580>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00035354867577552795,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "text_then_weather_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07129869516938925,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.5799349313601851,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-text_then_weather_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f427410c580>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'text_then_weather_tool', 'expected': [{'answer': ['sol'], 'num_tool_calls': 0}, {'num_tool_calls': 1, 'to...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f427417b3a0>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00033699069172143936,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "weather_tool_then_text"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07074506860226393,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.5245106862857938,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-weather_tool_then_text]>>\nopenai_client = <openai.OpenAI object at 0x7f427430e590>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'weather_tool_then_text', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'location': 'San Francisco...], 'type': 'object'}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': '70 degrees and foggy'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f4274268a90>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00042015407234430313,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "add_product_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07020766660571098,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.6389470677822828,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-add_product_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f42741784f0>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'add_product_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'inStock': True, 'name': 'Widget...}}, 'type': 'function'}]}, 'tool_responses': [{'response': \"{'response': 'Successfully added product with id: 123'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f4274254bb0>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00035757478326559067,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "get_then_create_event_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07121358439326286,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.5222592242062092,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-get_then_create_event_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f42741e8ca0>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'get_then_create_event_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'date': '2025-03-03', ...ents found for 2025-03-03 at 10:00'}\"}, {'response': \"{'response': 'Successfully created new event with id: e_123'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f427416c6a0>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003436664119362831,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]",
|
|
"lineno": 471,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "compare_monthly_expense_tool"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07017400953918695,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 1.7245550760999322,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 506,
|
|
"message": ""
|
|
},
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 688,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_streaming_multi_turn_tool_calling[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-compare_monthly_expense_tool]>>\nopenai_client = <openai.OpenAI object at 0x7f4274256b90>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\ncase = {'case_id': 'compare_monthly_expense_tool', 'expected': [{'num_tool_calls': 1, 'tool_arguments': {'month': 1, 'year': ... 'Total expenses for January 2025: $1000'}\"}, {'response': \"{'response': 'Total expenses for February 2024: $2000'}\"}]}\n\n @pytest.mark.parametrize(\n \"case\",\n chat_completion_test_cases.get(\"test_chat_multi_turn_tool_calling\", {}).get(\"test_params\", {}).get(\"case\", []),\n ids=case_id_generator,\n )\n def test_chat_streaming_multi_turn_tool_calling(request, openai_client, model, provider, verification_config, case):\n \"\"\" \"\"\"\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages = []\n tools = case[\"input\"][\"tools\"]\n expected_results = copy.deepcopy(case[\"expected\"])\n tool_responses = copy.deepcopy(case.get(\"tool_responses\", []))\n input_messages_turns = copy.deepcopy(case[\"input\"][\"messages\"])\n \n while len(input_messages_turns) > 0 or (len(messages) > 0 and messages[-1][\"role\"] == \"tool\"):\n if len(messages) == 0 or messages[-1][\"role\"] != \"tool\":\n new_messages = input_messages_turns.pop(0)\n if isinstance(new_messages, list):\n messages.extend(new_messages)\n else:\n messages.append(new_messages)\n \n # --- API Call (Streaming) ---\n stream = openai_client.chat.completions.create(\n model=model,\n messages=messages,\n tools=tools,\n stream=True,\n )\n \n # --- Process Stream ---\n> accumulated_content, accumulated_tool_calls = _accumulate_streaming_tool_calls(stream)\n\ntests/verifications/openai_api/test_chat_completion.py:506: \n_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \n\nstream = <openai.Stream object at 0x7f427415f0a0>\n\n def _accumulate_streaming_tool_calls(stream):\n \"\"\"Accumulates tool calls and content from a streaming ChatCompletion response.\"\"\"\n tool_calls_buffer = {}\n current_id = None\n full_content = \"\" # Initialize content accumulator\n # Process streaming chunks\n for chunk in stream:\n> choice = chunk.choices[0]\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:688: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003162780776619911,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=False]",
|
|
"lineno": 554,
|
|
"outcome": "skipped",
|
|
"keywords": [
|
|
"test_chat_multi_turn_multiple_images[meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=False]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=False",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "stream=False"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07253758516162634,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.00021537486463785172,
|
|
"outcome": "skipped",
|
|
"longrepr": "('/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py', 561, 'Skipped: Skipping test_chat_multi_turn_multiple_images for model meta-llama/Llama-3.3-70B-Instruct-Turbo on provider together based on config.')"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0004162406548857689,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=True]",
|
|
"lineno": 554,
|
|
"outcome": "skipped",
|
|
"keywords": [
|
|
"test_chat_multi_turn_multiple_images[meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=True]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-3.3-70B-Instruct-Turbo-stream=True",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"case_id": "stream=True"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07268107868731022,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 0.0002132616937160492,
|
|
"outcome": "skipped",
|
|
"longrepr": "('/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py', 561, 'Skipped: Skipping test_chat_multi_turn_multiple_images for model meta-llama/Llama-3.3-70B-Instruct-Turbo on provider together based on config.')"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.00021094270050525665,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=False]",
|
|
"lineno": 554,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=False]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=False",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "stream=False"
|
|
},
|
|
"setup": {
|
|
"duration": 0.07398672867566347,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 4.383559702895582,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002781357616186142,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=True]",
|
|
"lineno": 554,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=True]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=True",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
"case_id": "stream=True"
|
|
},
|
|
"setup": {
|
|
"duration": 0.08006586041301489,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.16784877050668,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 596,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 596,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Scout-17B-16E-Instruct-stream=True]>>\nopenai_client = <openai.OpenAI object at 0x7f427416c490>\nmodel = 'meta-llama/Llama-4-Scout-17B-16E-Instruct', provider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\nmulti_image_data = ['data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGC...6pH9jaTzNv7vfRRXzubfxj9f8Pv8AkTz/AMX/ALbEz5Ly38lfMk/5Z/u64PxhqEZh+z/6rzvn2UUV5EvgPuzy/wAc6p5dt5ccibJpNkkdFFFec27mZ//Z']\nstream = True\n\n @pytest.mark.parametrize(\"stream\", [False, True], ids=[\"stream=False\", \"stream=True\"])\n def test_chat_multi_turn_multiple_images(\n request, openai_client, model, provider, verification_config, multi_image_data, stream\n ):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages_turn1 = [\n {\n \"role\": \"user\",\n \"content\": [\n {\n \"type\": \"image_url\",\n \"image_url\": {\n \"url\": multi_image_data[0],\n },\n },\n {\n \"type\": \"image_url\",\n \"image_url\": {\n \"url\": multi_image_data[1],\n },\n },\n {\n \"type\": \"text\",\n \"text\": \"What furniture is in the first image that is not in the second image?\",\n },\n ],\n },\n ]\n \n # First API call\n response1 = openai_client.chat.completions.create(\n model=model,\n messages=messages_turn1,\n stream=stream,\n )\n if stream:\n message_content1 = \"\"\n for chunk in response1:\n> message_content1 += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:596: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0003619194030761719,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=False]",
|
|
"lineno": 554,
|
|
"outcome": "passed",
|
|
"keywords": [
|
|
"test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=False]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=False",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "stream=False"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0709412069991231,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 6.110534753650427,
|
|
"outcome": "passed"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0002450142055749893,
|
|
"outcome": "passed"
|
|
}
|
|
},
|
|
{
|
|
"nodeid": "tests/verifications/openai_api/test_chat_completion.py::test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=True]",
|
|
"lineno": 554,
|
|
"outcome": "failed",
|
|
"keywords": [
|
|
"test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=True]",
|
|
"parametrize",
|
|
"pytestmark",
|
|
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=True",
|
|
"test_chat_completion.py",
|
|
"openai_api",
|
|
"verifications",
|
|
"tests",
|
|
"llama-stack",
|
|
""
|
|
],
|
|
"metadata": {
|
|
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
"case_id": "stream=True"
|
|
},
|
|
"setup": {
|
|
"duration": 0.0725309094414115,
|
|
"outcome": "passed"
|
|
},
|
|
"call": {
|
|
"duration": 2.291131243109703,
|
|
"outcome": "failed",
|
|
"crash": {
|
|
"path": "/home/erichuang/llama-stack/tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 596,
|
|
"message": "IndexError: list index out of range"
|
|
},
|
|
"traceback": [
|
|
{
|
|
"path": "tests/verifications/openai_api/test_chat_completion.py",
|
|
"lineno": 596,
|
|
"message": "IndexError"
|
|
}
|
|
],
|
|
"longrepr": "request = <FixtureRequest for <Function test_chat_multi_turn_multiple_images[meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8-stream=True]>>\nopenai_client = <openai.OpenAI object at 0x7f42740eb0d0>\nmodel = 'meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'\nprovider = 'together'\nverification_config = {'providers': {'cerebras': {'api_key_var': 'CEREBRAS_API_KEY', 'base_url': 'https://api.cerebras.ai/v1', 'model_displa...-versatile', 'meta-llama/llama-4-scout-17b-16e-instruct', 'meta-llama/llama-4-maverick-17b-128e-instruct'], ...}, ...}}\nmulti_image_data = ['data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGC...6pH9jaTzNv7vfRRXzubfxj9f8Pv8AkTz/AMX/ALbEz5Ly38lfMk/5Z/u64PxhqEZh+z/6rzvn2UUV5EvgPuzy/wAc6p5dt5ccibJpNkkdFFFec27mZ//Z']\nstream = True\n\n @pytest.mark.parametrize(\"stream\", [False, True], ids=[\"stream=False\", \"stream=True\"])\n def test_chat_multi_turn_multiple_images(\n request, openai_client, model, provider, verification_config, multi_image_data, stream\n ):\n test_name_base = get_base_test_name(request)\n if should_skip_test(verification_config, provider, model, test_name_base):\n pytest.skip(f\"Skipping {test_name_base} for model {model} on provider {provider} based on config.\")\n \n messages_turn1 = [\n {\n \"role\": \"user\",\n \"content\": [\n {\n \"type\": \"image_url\",\n \"image_url\": {\n \"url\": multi_image_data[0],\n },\n },\n {\n \"type\": \"image_url\",\n \"image_url\": {\n \"url\": multi_image_data[1],\n },\n },\n {\n \"type\": \"text\",\n \"text\": \"What furniture is in the first image that is not in the second image?\",\n },\n ],\n },\n ]\n \n # First API call\n response1 = openai_client.chat.completions.create(\n model=model,\n messages=messages_turn1,\n stream=stream,\n )\n if stream:\n message_content1 = \"\"\n for chunk in response1:\n> message_content1 += chunk.choices[0].delta.content or \"\"\nE IndexError: list index out of range\n\ntests/verifications/openai_api/test_chat_completion.py:596: IndexError"
|
|
},
|
|
"teardown": {
|
|
"duration": 0.0018906639888882637,
|
|
"outcome": "passed"
|
|
}
|
|
}
|
|
],
|
|
"run_timestamp": 1744918065
|
|
}
|