mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-10-14 22:33:48 +00:00
045a0c1d57
Uses test_id in request hashes and test-scoped subdirectories to prevent cross-test contamination. Model list endpoints exclude test_id to enable merging recordings from different servers. Additionally, this PR adds a `record-if-missing` mode (which we will use instead of `record` which records everything) which is very useful. 🤖 Co-authored with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude <noreply@anthropic.com>
57 lines
2.7 KiB
JSON
57 lines
2.7 KiB
JSON
{
|
|
"test_id": null,
|
|
"request": {
|
|
"method": "POST",
|
|
"url": "http://0.0.0.0:11434/v1/v1/chat/completions",
|
|
"headers": {},
|
|
"body": {
|
|
"model": "llama3.2:3b-instruct-fp16",
|
|
"messages": [
|
|
{
|
|
"role": "user",
|
|
"content": "Test trace openai 2"
|
|
}
|
|
],
|
|
"stream": false
|
|
},
|
|
"endpoint": "/v1/chat/completions",
|
|
"model": "llama3.2:3b-instruct-fp16"
|
|
},
|
|
"response": {
|
|
"body": {
|
|
"__type__": "openai.types.chat.chat_completion.ChatCompletion",
|
|
"__data__": {
|
|
"id": "rec-395c30078677",
|
|
"choices": [
|
|
{
|
|
"finish_reason": "stop",
|
|
"index": 0,
|
|
"logprobs": null,
|
|
"message": {
|
|
"content": "I'd be happy to help you test the OpenAI 2.0 model!\n\nPlease note that I need to clarify a few things:\n\n1. **I can only simulate the conversation**: I won't have direct access to the actual OpenAI 2.0 model.\n2. **The simulation might not be perfect**: The testing process will be based on my understanding of the model's capabilities and limitations.\n\nThat being said, here's how you can test the OpenAI 2.0 model with me:\n\n**Testing Scenarios**\n\nChoose one or more of the following scenarios to test the OpenAI 2.0 model:\n\n1. **Basic Conversation**: Test basic conversations, such as asking questions and receiving answers.\n2. **Dialogue Flow**: Test the model's ability to maintain a conversation flow by asking follow-up questions.\n3. **Creative Writing**: Test the model's creativity by asking it to generate paragraphs or short stories.\n4. **Fact Retrieval**: Test the model's knowledge retrieval capabilities by asking it factual questions.\n\n**Input Format**\n\nPlease format your input using plain text. You can use:\n\n* A single question or statement\n* Multiple statements separated by commas\n* A prompt with a specific format (e.g., \"Write a short story about [topic]\")\n\n**Example Input**\n```\nAsk me: What is the capital of France?\n```\n\nGo ahead and choose your testing scenario, and I'll simulate the conversation!",
|
|
"refusal": null,
|
|
"role": "assistant",
|
|
"annotations": null,
|
|
"audio": null,
|
|
"function_call": null,
|
|
"tool_calls": null
|
|
}
|
|
}
|
|
],
|
|
"created": 0,
|
|
"model": "llama3.2:3b-instruct-fp16",
|
|
"object": "chat.completion",
|
|
"service_tier": null,
|
|
"system_fingerprint": "fp_ollama",
|
|
"usage": {
|
|
"completion_tokens": 293,
|
|
"prompt_tokens": 31,
|
|
"total_tokens": 324,
|
|
"completion_tokens_details": null,
|
|
"prompt_tokens_details": null
|
|
}
|
|
}
|
|
},
|
|
"is_streaming": false
|
|
}
|
|
}
|