phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-22 20:40:00 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

improvements to include other ollama methods like ps, list and pull

Browse source
This commit is contained in:
Ashwin Bharambe 2025-07-29 10:56:57 -07:00
parent 1a21c4b695
commit b578f9aec1
23 changed files with 6748 additions and 5711 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama_stack/distribution/routers/inference.py
View file

@ -77,13 +77,13 @@ class InferenceRouter(Inference):
self.tokenizer = Tokenizer.get_instance() self.tokenizer = Tokenizer.get_instance()
self.formatter = ChatFormat(self.tokenizer) self.formatter = ChatFormat(self.tokenizer)
self.recording_context = None
async def initialize(self) -> None: async def initialize(self) -> None:
logger.debug("InferenceRouter.initialize") logger.debug("InferenceRouter.initialize")
pass
async def shutdown(self) -> None: async def shutdown(self) -> None:
logger.debug("InferenceRouter.shutdown") logger.debug("InferenceRouter.shutdown")
pass
async def register_model( async def register_model(
self, self,

17
llama_stack/distribution/stack.py
View file

@ -94,6 +94,7 @@ RESOURCES = [
REGISTRY_REFRESH_INTERVAL_SECONDS = 300 REGISTRY_REFRESH_INTERVAL_SECONDS = 300
REGISTRY_REFRESH_TASK = None REGISTRY_REFRESH_TASK = None
TEST_RECORDING_CONTEXT = None
async def register_resources(run_config: StackRunConfig, impls: dict[Api, Any]): async def register_resources(run_config: StackRunConfig, impls: dict[Api, Any]):
@ -307,6 +308,15 @@ def add_internal_implementations(impls: dict[Api, Any], run_config: StackRunConf
async def construct_stack( async def construct_stack(
run_config: StackRunConfig, provider_registry: ProviderRegistry | None = None run_config: StackRunConfig, provider_registry: ProviderRegistry | None = None
) -> dict[Api, Any]: ) -> dict[Api, Any]:
inference_mode = os.environ.get("LLAMA_STACK_TEST_INFERENCE_MODE", "live").lower()
if inference_mode in ["record", "replay"]:
global TEST_RECORDING_CONTEXT
from llama_stack.testing.inference_recorder import setup_inference_recording
TEST_RECORDING_CONTEXT = setup_inference_recording()
TEST_RECORDING_CONTEXT.__enter__()
logger.info(f"Inference recording enabled: mode={inference_mode}")
dist_registry, _ = await create_dist_registry(run_config.metadata_store, run_config.image_name) dist_registry, _ = await create_dist_registry(run_config.metadata_store, run_config.image_name)
policy = run_config.server.auth.access_policy if run_config.server.auth else [] policy = run_config.server.auth.access_policy if run_config.server.auth else []
impls = await resolve_impls( impls = await resolve_impls(
@ -352,6 +362,13 @@ async def shutdown_stack(impls: dict[Api, Any]):
except (Exception, asyncio.CancelledError) as e: except (Exception, asyncio.CancelledError) as e:
logger.exception(f"Failed to shutdown {impl_name}: {e}") logger.exception(f"Failed to shutdown {impl_name}: {e}")
global TEST_RECORDING_CONTEXT
if TEST_RECORDING_CONTEXT:
try:
TEST_RECORDING_CONTEXT.__exit__(None, None, None)
except Exception as e:
logger.error(f"Error during inference recording cleanup: {e}")
global REGISTRY_REFRESH_TASK global REGISTRY_REFRESH_TASK
if REGISTRY_REFRESH_TASK: if REGISTRY_REFRESH_TASK:
REGISTRY_REFRESH_TASK.cancel() REGISTRY_REFRESH_TASK.cancel()

70
llama_stack/testing/inference_recorder.py
View file

@ -13,13 +13,23 @@ import sqlite3
from collections.abc import Generator from collections.abc import Generator
from contextlib import contextmanager from contextlib import contextmanager
from pathlib import Path from pathlib import Path
from typing import Any, cast from typing import Any, Literal, cast
from llama_stack.log import get_logger
logger = get_logger(__name__, category="testing")
# Global state for the recording system # Global state for the recording system
_current_mode: str | None = None _current_mode: str | None = None
_current_storage: ResponseStorage | None = None _current_storage: ResponseStorage | None = None
_original_methods: dict[str, Any] = {} _original_methods: dict[str, Any] = {}
from openai.types.completion_choice import CompletionChoice
# update the "finish_reason" field, since its type definition is wrong (no None is accepted)
CompletionChoice.model_fields["finish_reason"].annotation = Literal["stop", "length", "content_filter"] | None
CompletionChoice.model_rebuild()
def normalize_request(method: str, url: str, headers: dict[str, Any], body: dict[str, Any]) -> str: def normalize_request(method: str, url: str, headers: dict[str, Any], body: dict[str, Any]) -> str:
"""Create a normalized hash of the request for consistent matching.""" """Create a normalized hash of the request for consistent matching."""
@ -35,14 +45,14 @@ def normalize_request(method: str, url: str, headers: dict[str, Any], body: dict
def get_inference_mode() -> str: def get_inference_mode() -> str:
return os.environ.get("LLAMA_STACK_INFERENCE_MODE", "live").lower() return os.environ.get("LLAMA_STACK_TEST_INFERENCE_MODE", "live").lower()
def setup_inference_recording(): def setup_inference_recording():
mode = get_inference_mode() mode = get_inference_mode()
if mode not in ["live", "record", "replay"]: if mode not in ["live", "record", "replay"]:
raise ValueError(f"Invalid LLAMA_STACK_INFERENCE_MODE: {mode}. Must be 'live', 'record', or 'replay'") raise ValueError(f"Invalid LLAMA_STACK_TEST_INFERENCE_MODE: {mode}. Must be 'live', 'record', or 'replay'")
if mode == "live": if mode == "live":
# Return a no-op context manager for live mode # Return a no-op context manager for live mode
@ -52,23 +62,43 @@ def setup_inference_recording():
return live_mode() return live_mode()
if "LLAMA_STACK_RECORDING_DIR" not in os.environ: if "LLAMA_STACK_TEST_RECORDING_DIR" not in os.environ:
raise ValueError("LLAMA_STACK_RECORDING_DIR must be set for recording or replaying") raise ValueError("LLAMA_STACK_TEST_RECORDING_DIR must be set for recording or replaying")
storage_dir = os.environ["LLAMA_STACK_RECORDING_DIR"] storage_dir = os.environ["LLAMA_STACK_TEST_RECORDING_DIR"]
return inference_recording(mode=mode, storage_dir=storage_dir) return inference_recording(mode=mode, storage_dir=storage_dir)
def _serialize_response(response: Any) -> Any: def _serialize_response(response: Any) -> Any:
if hasattr(response, "model_dump"): if hasattr(response, "model_dump"):
return response.model_dump() data = response.model_dump(mode="json")
return {
"__type__": f"{response.__class__.__module__}.{response.__class__.__qualname__}",
"__data__": data,
}
elif hasattr(response, "__dict__"): elif hasattr(response, "__dict__"):
return dict(response.__dict__) return dict(response.__dict__)
else: else:
return response return response
def _deserialize_response(data: dict[str, Any]) -> dict[str, Any]: def _deserialize_response(data: dict[str, Any]) -> Any:
# Check if this is a serialized Pydantic model with type information
if isinstance(data, dict) and "__type__" in data and "__data__" in data:
try:
# Import the original class and reconstruct the object
module_path, class_name = data["__type__"].rsplit(".", 1)
module = __import__(module_path, fromlist=[class_name])
cls = getattr(module, class_name)
if not hasattr(cls, "model_validate"):
raise ValueError(f"Pydantic class {cls} does not support model_validate?")
return cls.model_validate(data["__data__"])
except (ImportError, AttributeError, TypeError, ValueError) as e:
logger.warning(f"Failed to deserialize object of type {data['__type__']}: {e}")
return data["__data__"]
return data return data
@ -120,6 +150,7 @@ class ResponseStorage:
with open(response_path, "w") as f: with open(response_path, "w") as f:
json.dump({"request": request, "response": serialized_response}, f, indent=2) json.dump({"request": request, "response": serialized_response}, f, indent=2)
f.write("\n") f.write("\n")
f.flush()
# Update SQLite index # Update SQLite index
with sqlite3.connect(self.db_path) as conn: with sqlite3.connect(self.db_path) as conn:
@ -214,6 +245,8 @@ async def _patched_inference_method(original_method, self, client_type, method_n
endpoint = "/api/chat" endpoint = "/api/chat"
elif method_name == "embed": elif method_name == "embed":
endpoint = "/api/embeddings" endpoint = "/api/embeddings"
elif method_name == "list":
endpoint = "/api/tags"
else: else:
endpoint = f"/api/{method_name}" endpoint = f"/api/{method_name}"
else: else:
@ -295,8 +328,6 @@ def patch_inference_clients():
"""Install monkey patches for OpenAI client methods and Ollama AsyncClient methods.""" """Install monkey patches for OpenAI client methods and Ollama AsyncClient methods."""
global _original_methods global _original_methods
# Import here to avoid circular imports
# Also import Ollama AsyncClient
from ollama import AsyncClient as OllamaAsyncClient from ollama import AsyncClient as OllamaAsyncClient
from openai.resources.chat.completions import AsyncCompletions as AsyncChatCompletions from openai.resources.chat.completions import AsyncCompletions as AsyncChatCompletions
from openai.resources.completions import AsyncCompletions from openai.resources.completions import AsyncCompletions
@ -310,6 +341,9 @@ def patch_inference_clients():
"ollama_generate": OllamaAsyncClient.generate, "ollama_generate": OllamaAsyncClient.generate,
"ollama_chat": OllamaAsyncClient.chat, "ollama_chat": OllamaAsyncClient.chat,
"ollama_embed": OllamaAsyncClient.embed, "ollama_embed": OllamaAsyncClient.embed,
"ollama_ps": OllamaAsyncClient.ps,
"ollama_pull": OllamaAsyncClient.pull,
"ollama_list": OllamaAsyncClient.list,
} }
# Create patched methods for OpenAI client # Create patched methods for OpenAI client
@ -339,10 +373,24 @@ def patch_inference_clients():
async def patched_ollama_embed(self, **kwargs): async def patched_ollama_embed(self, **kwargs):
return await _patched_inference_method(_original_methods["ollama_embed"], self, "ollama", "embed", **kwargs) return await _patched_inference_method(_original_methods["ollama_embed"], self, "ollama", "embed", **kwargs)
async def patched_ollama_ps(self, **kwargs):
logger.info("replay mode: ollama.ps() reporting success")
return []
async def patched_ollama_pull(self, *args, **kwargs):
logger.info("replay mode: ollama.pull() not actually pulling the model")
return None
async def patched_ollama_list(self, **kwargs):
return await _patched_inference_method(_original_methods["ollama_list"], self, "ollama", "list", **kwargs)
# Apply Ollama patches # Apply Ollama patches
OllamaAsyncClient.generate = patched_ollama_generate OllamaAsyncClient.generate = patched_ollama_generate
OllamaAsyncClient.chat = patched_ollama_chat OllamaAsyncClient.chat = patched_ollama_chat
OllamaAsyncClient.embed = patched_ollama_embed OllamaAsyncClient.embed = patched_ollama_embed
OllamaAsyncClient.ps = patched_ollama_ps
OllamaAsyncClient.pull = patched_ollama_pull
OllamaAsyncClient.list = patched_ollama_list
def unpatch_inference_clients(): def unpatch_inference_clients():
@ -367,6 +415,8 @@ def unpatch_inference_clients():
OllamaAsyncClient.generate = _original_methods["ollama_generate"] OllamaAsyncClient.generate = _original_methods["ollama_generate"]
OllamaAsyncClient.chat = _original_methods["ollama_chat"] OllamaAsyncClient.chat = _original_methods["ollama_chat"]
OllamaAsyncClient.embed = _original_methods["ollama_embed"] OllamaAsyncClient.embed = _original_methods["ollama_embed"]
OllamaAsyncClient.ps = _original_methods["ollama_ps"]
OllamaAsyncClient.list = _original_methods["ollama_list"]
_original_methods.clear() _original_methods.clear()

54
tests/integration/fixtures/common.py
View file

@ -185,18 +185,6 @@ def llama_stack_client(request, provider_data):
if not config: if not config:
raise ValueError("You must specify either --stack-config or LLAMA_STACK_CONFIG") raise ValueError("You must specify either --stack-config or LLAMA_STACK_CONFIG")
# Set up inference recording if enabled
inference_mode = os.environ.get("LLAMA_STACK_INFERENCE_MODE", "live").lower()
recording_context = None
if inference_mode in ["record", "replay"]:
from llama_stack.testing.inference_recorder import setup_inference_recording
recording_context = setup_inference_recording()
recording_context.__enter__()
print(f"Inference recording enabled: mode={inference_mode}")
try:
# Handle server:<config_name> format or server:<config_name>:<port> # Handle server:<config_name> format or server:<config_name>:<port>
if config.startswith("server:"): if config.startswith("server:"):
parts = config.split(":") parts = config.split(":")
@ -227,24 +215,24 @@ def llama_stack_client(request, provider_data):
else: else:
print(f"Port {port} is already in use, assuming server is already running...") print(f"Port {port} is already in use, assuming server is already running...")
client = LlamaStackClient( return LlamaStackClient(
base_url=base_url, base_url=base_url,
provider_data=provider_data, provider_data=provider_data,
timeout=int(os.environ.get("LLAMA_STACK_CLIENT_TIMEOUT", "30")), timeout=int(os.environ.get("LLAMA_STACK_CLIENT_TIMEOUT", "30")),
) )
else:
# check if this looks like a URL using proper URL parsing # check if this looks like a URL using proper URL parsing
try: try:
parsed_url = urlparse(config) parsed_url = urlparse(config)
if parsed_url.scheme and parsed_url.netloc: if parsed_url.scheme and parsed_url.netloc:
client = LlamaStackClient( return LlamaStackClient(
base_url=config, base_url=config,
provider_data=provider_data, provider_data=provider_data,
) )
else: except Exception:
raise ValueError("Not a URL") # If URL parsing fails, treat as non-URL config
except Exception as e: pass
# If URL parsing fails, treat as library config
if "=" in config: if "=" in config:
run_config = run_config_from_adhoc_config_spec(config) run_config = run_config_from_adhoc_config_spec(config)
run_config_file = tempfile.NamedTemporaryFile(delete=False, suffix=".yaml") run_config_file = tempfile.NamedTemporaryFile(delete=False, suffix=".yaml")
@ -258,23 +246,10 @@ def llama_stack_client(request, provider_data):
skip_logger_removal=True, skip_logger_removal=True,
) )
if not client.initialize(): if not client.initialize():
raise RuntimeError("Initialization failed") from e raise RuntimeError("Initialization failed")
# Store recording context for cleanup
if recording_context:
request.session._inference_recording_context = recording_context
return client return client
except Exception:
# Clean up recording context on error
if recording_context:
try:
recording_context.__exit__(None, None, None)
except Exception as cleanup_error:
print(f"Warning: Error cleaning up recording context: {cleanup_error}")
raise
@pytest.fixture(scope="session") @pytest.fixture(scope="session")
def openai_client(client_with_models): def openai_client(client_with_models):
@ -289,20 +264,9 @@ def compat_client(request):
@pytest.fixture(scope="session", autouse=True) @pytest.fixture(scope="session", autouse=True)
def cleanup_server_process(request): def cleanup_server_process(request):
"""Cleanup server process and inference recording at the end of the test session.""" """Cleanup server process at the end of the test session."""
yield # Run tests yield # Run tests
# Clean up inference recording context
if hasattr(request.session, "_inference_recording_context"):
recording_context = request.session._inference_recording_context
if recording_context:
try:
print("Cleaning up inference recording context...")
recording_context.__exit__(None, None, None)
except Exception as e:
print(f"Error during inference recording cleanup: {e}")
# Clean up server process
if hasattr(request.session, "_llama_stack_server_process"): if hasattr(request.session, "_llama_stack_server_process"):
server_process = request.session._llama_stack_server_process server_process = request.session._llama_stack_server_process
if server_process: if server_process:

BIN
tests/integration/inference/recordings/index.sqlite
View file

Binary file not shown.

13
tests/integration/inference/recordings/responses/1b8394f90636.json
View file

@ -19,19 +19,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.555099Z", "created_at": "2025-07-29T17:37:01.657977Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 2124168875, "total_duration": 1860746958,
"load_duration": 58506875, "load_duration": 59632791,
"prompt_eval_count": 18, "prompt_eval_count": 18,
"prompt_eval_duration": 70072583, "prompt_eval_duration": 66333291,
"eval_count": 43, "eval_count": 43,
"eval_duration": 1994446917, "eval_duration": 1734228875,
"response": " _______.\n\nThe best answer is blue. The traditional nursery rhyme goes like this:\n\nRoses are red,\nViolets are blue,\nSugar is sweet,\nAnd so are you! (Or something similar.)", "response": " _______.\n\nThe best answer is blue. The traditional nursery rhyme goes like this:\n\nRoses are red,\nViolets are blue,\nSugar is sweet,\nAnd so are you! (Or something similar.)",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

13
tests/integration/inference/recordings/responses/211b1562d4e6.json
View file

@ -17,19 +17,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.535525Z", "created_at": "2025-07-29T17:37:05.658757Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 358691334, "total_duration": 327636667,
"load_duration": 76787334, "load_duration": 55751292,
"prompt_eval_count": 23, "prompt_eval_count": 23,
"prompt_eval_duration": 72235375, "prompt_eval_duration": 65865625,
"eval_count": 6, "eval_count": 6,
"eval_duration": 208986666, "eval_duration": 205366667,
"response": "Humans live on Earth.", "response": "Humans live on Earth.",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

63
tests/integration/inference/recordings/responses/3c3f13cb7794.json
View file

@ -18,8 +18,10 @@
"response": { "response": {
"body": [ "body": [
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.691771Z", "created_at": "2025-07-29T17:37:05.833193Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -31,10 +33,13 @@
"response": "The", "response": "The",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.732262Z", "created_at": "2025-07-29T17:37:05.874274Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -46,10 +51,13 @@
"response": " Latin", "response": " Latin",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.77294Z", "created_at": "2025-07-29T17:37:05.915195Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -61,10 +69,13 @@
"response": " word", "response": " word",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.814484Z", "created_at": "2025-07-29T17:37:05.955964Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -76,10 +87,13 @@
"response": " for", "response": " for",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.854875Z", "created_at": "2025-07-29T17:37:05.996679Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -91,10 +105,13 @@
"response": " \"", "response": " \"",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.895957Z", "created_at": "2025-07-29T17:37:06.037268Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -106,10 +123,13 @@
"response": "Sun", "response": "Sun",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.937445Z", "created_at": "2025-07-29T17:37:06.078124Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -121,10 +141,13 @@
"response": "\"", "response": "\"",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.978832Z", "created_at": "2025-07-29T17:37:06.118771Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -136,10 +159,13 @@
"response": " is", "response": " is",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:58.019242Z", "created_at": "2025-07-29T17:37:06.159758Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -151,10 +177,13 @@
"response": " Sol", "response": " Sol",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:58.059902Z", "created_at": "2025-07-29T17:37:06.200659Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -166,22 +195,26 @@
"response": ".", "response": ".",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:58.100535Z", "created_at": "2025-07-29T17:37:06.241464Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 528254250, "total_duration": 542773583,
"load_duration": 50177125, "load_duration": 68681541,
"prompt_eval_count": 26, "prompt_eval_count": 26,
"prompt_eval_duration": 68018458, "prompt_eval_duration": 64427833,
"eval_count": 11, "eval_count": 11,
"eval_duration": 409555959, "eval_duration": 409175667,
"response": "", "response": "",
"thinking": null, "thinking": null,
"context": null "context": null
} }
}
], ],
"is_streaming": true "is_streaming": true
} }

63
tests/integration/inference/recordings/responses/40f524d1934a.json
View file

@ -18,8 +18,10 @@
"response": { "response": {
"body": [ "body": [
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.480955Z", "created_at": "2025-07-29T17:37:07.590293Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -31,10 +33,13 @@
"response": "[", "response": "[",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.527418Z", "created_at": "2025-07-29T17:37:07.638879Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -46,10 +51,13 @@
"response": "get", "response": "get",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.571522Z", "created_at": "2025-07-29T17:37:07.69047Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -61,10 +69,13 @@
"response": "_weather", "response": "_weather",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.615027Z", "created_at": "2025-07-29T17:37:07.738247Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -76,10 +87,13 @@
"response": "(location", "response": "(location",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.660598Z", "created_at": "2025-07-29T17:37:07.782316Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -91,10 +105,13 @@
"response": "=\"", "response": "=\"",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.705052Z", "created_at": "2025-07-29T17:37:07.827405Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -106,10 +123,13 @@
"response": "San", "response": "San",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.754386Z", "created_at": "2025-07-29T17:37:07.872241Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -121,10 +141,13 @@
"response": " Francisco", "response": " Francisco",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.796942Z", "created_at": "2025-07-29T17:37:07.91705Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -136,10 +159,13 @@
"response": ",", "response": ",",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.845807Z", "created_at": "2025-07-29T17:37:07.962046Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -151,10 +177,13 @@
"response": " CA", "response": " CA",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.891254Z", "created_at": "2025-07-29T17:37:08.00691Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -166,22 +195,26 @@
"response": "\")]", "response": "\")]",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.934197Z", "created_at": "2025-07-29T17:37:08.052731Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 574307083, "total_duration": 649504167,
"load_duration": 72062083, "load_duration": 138022250,
"prompt_eval_count": 324, "prompt_eval_count": 324,
"prompt_eval_duration": 47115625, "prompt_eval_duration": 45176916,
"eval_count": 11, "eval_count": 11,
"eval_duration": 454426708, "eval_duration": 464930417,
"response": "", "response": "",
"thinking": null, "thinking": null,
"context": null "context": null
} }
}
], ],
"is_streaming": true "is_streaming": true
} }

13
tests/integration/inference/recordings/responses/48d2fb183a2a.json
View file

@ -64,19 +64,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:13.438182Z", "created_at": "2025-07-29T17:37:21.486763Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 2975265833, "total_duration": 2935272125,
"load_duration": 95592083, "load_duration": 51690583,
"prompt_eval_count": 259, "prompt_eval_count": 259,
"prompt_eval_duration": 367103709, "prompt_eval_duration": 369929333,
"eval_count": 60, "eval_count": 60,
"eval_duration": 2511576708, "eval_duration": 2512928125,
"response": "{\n \"first_name\": \"Michael\",\n \"last_name\": \"Jordan\",\n \"year_of_birth\": 1963,\n \"nba_stats\": {\n \"year_for_draft\": 1984,\n \"num_seasons_in_nba\": 15\n }\n}", "response": "{\n \"first_name\": \"Michael\",\n \"last_name\": \"Jordan\",\n \"year_of_birth\": 1963,\n \"nba_stats\": {\n \"year_for_draft\": 1984,\n \"num_seasons_in_nba\": 15\n }\n}",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

132
tests/integration/inference/recordings/responses/4a3a4447b16b.json Normal file
View file

@ -0,0 +1,132 @@
{
"request": {
"method": "POST",
"url": "http://localhost:11434/api/tags",
"headers": {},
"body": {},
"endpoint": "/api/tags",
"model": ""
},
"response": {
"body": {
"__type__": "ollama._types.ListResponse",
"__data__": {
"models": [
{
"model": "nomic-embed-text:latest",
"modified_at": "2025-07-29T10:36:48.647829-07:00",
"digest": "0a109f422b47e3a30ba2b10eca18548e944e8a23073ee3f3e947efcf3c45e59f",
"size": 274302450,
"details": {
"parent_model": "",
"format": "gguf",
"family": "nomic-bert",
"families": [
"nomic-bert"
],
"parameter_size": "137M",
"quantization_level": "F16"
}
},
{
"model": "llama-guard3:1b",
"modified_at": "2025-07-25T14:39:44.978630-07:00",
"digest": "494147e06bf99e10dbe67b63a07ac81c162f18ef3341aa3390007ac828571b3b",
"size": 1600181919,
"details": {
"parent_model": "",
"format": "gguf",
"family": "llama",
"families": [
"llama"
],
"parameter_size": "1.5B",
"quantization_level": "Q8_0"
}
},
{
"model": "all-minilm:l6-v2",
"modified_at": "2025-07-24T15:15:11.129290-07:00",
"digest": "1b226e2802dbb772b5fc32a58f103ca1804ef7501331012de126ab22f67475ef",
"size": 45960996,
"details": {
"parent_model": "",
"format": "gguf",
"family": "bert",
"families": [
"bert"
],
"parameter_size": "23M",
"quantization_level": "F16"
}
},
{
"model": "llama3.2:1b",
"modified_at": "2025-07-17T22:02:24.953208-07:00",
"digest": "baf6a787fdffd633537aa2eb51cfd54cb93ff08e28040095462bb63daf552878",
"size": 1321098329,
"details": {
"parent_model": "",
"format": "gguf",
"family": "llama",
"families": [
"llama"
],
"parameter_size": "1.2B",
"quantization_level": "Q8_0"
}
},
{
"model": "all-minilm:latest",
"modified_at": "2025-06-03T16:50:10.946583-07:00",
"digest": "1b226e2802dbb772b5fc32a58f103ca1804ef7501331012de126ab22f67475ef",
"size": 45960996,
"details": {
"parent_model": "",
"format": "gguf",
"family": "bert",
"families": [
"bert"
],
"parameter_size": "23M",
"quantization_level": "F16"
}
},
{
"model": "llama3.2:3b",
"modified_at": "2025-05-01T11:15:23.797447-07:00",
"digest": "a80c4f17acd55265feec403c7aef86be0c25983ab279d83f3bcd3abbcb5b8b72",
"size": 2019393189,
"details": {
"parent_model": "",
"format": "gguf",
"family": "llama",
"families": [
"llama"
],
"parameter_size": "3.2B",
"quantization_level": "Q4_K_M"
}
},
{
"model": "llama3.2:3b-instruct-fp16",
"modified_at": "2025-04-30T15:33:48.939665-07:00",
"digest": "195a8c01d91ec3cb1e0aad4624a51f2602c51fa7d96110f8ab5a20c84081804d",
"size": 6433703586,
"details": {
"parent_model": "",
"format": "gguf",
"family": "llama",
"families": [
"llama"
],
"parameter_size": "3.2B",
"quantization_level": "F16"
}
}
]
}
},
"is_streaming": false
}
}

108
tests/integration/inference/recordings/responses/6cc063bbd7d3.json
View file

@ -18,8 +18,10 @@
"response": { "response": {
"body": [ "body": [
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.227427Z", "created_at": "2025-07-29T17:37:29.898943Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -31,10 +33,13 @@
"response": "The", "response": "The",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.275725Z", "created_at": "2025-07-29T17:37:29.940406Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -46,10 +51,13 @@
"response": " capital", "response": " capital",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.316195Z", "created_at": "2025-07-29T17:37:29.983928Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -61,10 +69,13 @@
"response": " of", "response": " of",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.356832Z", "created_at": "2025-07-29T17:37:30.025678Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -76,10 +87,13 @@
"response": " the", "response": " the",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.397682Z", "created_at": "2025-07-29T17:37:30.066554Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -91,10 +105,13 @@
"response": " United", "response": " United",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.438761Z", "created_at": "2025-07-29T17:37:30.111853Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -106,10 +123,13 @@
"response": " States", "response": " States",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.480453Z", "created_at": "2025-07-29T17:37:30.156263Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -121,10 +141,13 @@
"response": " is", "response": " is",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.523691Z", "created_at": "2025-07-29T17:37:30.197342Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -136,10 +159,13 @@
"response": " Washington", "response": " Washington",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.565106Z", "created_at": "2025-07-29T17:37:30.238939Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -151,10 +177,13 @@
"response": ",", "response": ",",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.606315Z", "created_at": "2025-07-29T17:37:30.28041Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -166,10 +195,13 @@
"response": " D", "response": " D",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.647209Z", "created_at": "2025-07-29T17:37:30.321152Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -181,10 +213,13 @@
"response": ".C", "response": ".C",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.687828Z", "created_at": "2025-07-29T17:37:30.362571Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -196,10 +231,13 @@
"response": ".", "response": ".",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.728386Z", "created_at": "2025-07-29T17:37:30.404107Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -211,10 +249,13 @@
"response": " (", "response": " (",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.769091Z", "created_at": "2025-07-29T17:37:30.444632Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -226,10 +267,13 @@
"response": "short", "response": "short",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.809726Z", "created_at": "2025-07-29T17:37:30.486331Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -241,10 +285,13 @@
"response": " for", "response": " for",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.850489Z", "created_at": "2025-07-29T17:37:30.527309Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -256,10 +303,13 @@
"response": " District", "response": " District",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.89147Z", "created_at": "2025-07-29T17:37:30.568556Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -271,10 +321,13 @@
"response": " of", "response": " of",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.932311Z", "created_at": "2025-07-29T17:37:30.610745Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -286,10 +339,13 @@
"response": " Columbia", "response": " Columbia",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:21.973566Z", "created_at": "2025-07-29T17:37:30.654172Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -301,22 +357,26 @@
"response": ").", "response": ").",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:22.014466Z", "created_at": "2025-07-29T17:37:30.695146Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 1034011167, "total_duration": 1078314875,
"load_duration": 176591709, "load_duration": 203057166,
"prompt_eval_count": 26, "prompt_eval_count": 26,
"prompt_eval_duration": 68104583, "prompt_eval_duration": 77142708,
"eval_count": 20, "eval_count": 20,
"eval_duration": 788670334, "eval_duration": 797458917,
"response": "", "response": "",
"thinking": null, "thinking": null,
"context": null "context": null
} }
}
], ],
"is_streaming": true "is_streaming": true
} }

13
tests/integration/inference/recordings/responses/70adef2c30c4.json
View file

@ -17,19 +17,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:20.924128Z", "created_at": "2025-07-29T17:37:29.59115Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 3308469666, "total_duration": 3560608959,
"load_duration": 66702250, "load_duration": 57756625,
"prompt_eval_count": 30, "prompt_eval_count": 30,
"prompt_eval_duration": 391410334, "prompt_eval_duration": 370892334,
"eval_count": 70, "eval_count": 70,
"eval_duration": 2849497291, "eval_duration": 3131360625,
"response": "The answer is Saturn! Saturn's ring system is one of the most iconic and well-known in our solar system. The rings are made up of ice particles, rock debris, and dust that orbit around the planet due to its gravitational pull.\n\nWould you like to know more about Saturn's rings or is there something else I can help you with?", "response": "The answer is Saturn! Saturn's ring system is one of the most iconic and well-known in our solar system. The rings are made up of ice particles, rock debris, and dust that orbit around the planet due to its gravitational pull.\n\nWould you like to know more about Saturn's rings or is there something else I can help you with?",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

13
tests/integration/inference/recordings/responses/75d0dd9d0fa3.json
View file

@ -42,19 +42,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:57.15491Z", "created_at": "2025-07-29T17:37:05.293829Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 1570055875, "total_duration": 1587647875,
"load_duration": 87677125, "load_duration": 58102458,
"prompt_eval_count": 119, "prompt_eval_count": 119,
"prompt_eval_duration": 190281458, "prompt_eval_duration": 199832792,
"eval_count": 29, "eval_count": 29,
"eval_duration": 1291217083, "eval_duration": 1328951417,
"response": "{ \"name\": \"Michael Jordan\", \"year_born\": \"1963\", \"year_retired\": \"2003\"}\n ", "response": "{ \"name\": \"Michael Jordan\", \"year_born\": \"1963\", \"year_retired\": \"2003\"}\n ",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

1006
tests/integration/inference/recordings/responses/84cab42e1f5c.json
View file

File diff suppressed because it is too large Load diff

13
tests/integration/inference/recordings/responses/9b812cbcb88d.json
View file

@ -17,19 +17,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:59.104609Z", "created_at": "2025-07-29T17:37:07.316473Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 948932208, "total_duration": 1001978333,
"load_duration": 68549542, "load_duration": 125002875,
"prompt_eval_count": 324, "prompt_eval_count": 324,
"prompt_eval_duration": 460136875, "prompt_eval_duration": 451915875,
"eval_count": 11, "eval_count": 11,
"eval_duration": 419553208, "eval_duration": 424435375,
"response": "[get_weather(location=\"San Francisco, CA\")]", "response": "[get_weather(location=\"San Francisco, CA\")]",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

13
tests/integration/inference/recordings/responses/9e7a83d3d596.json
View file

@ -13,26 +13,29 @@
}, },
"response": { "response": {
"body": { "body": {
"id": "cmpl-68", "__type__": "openai.types.completion.Completion",
"__data__": {
"id": "cmpl-256",
"choices": [ "choices": [
{ {
"finish_reason": "stop", "finish_reason": "stop",
"index": 0, "index": 0,
"logprobs": null, "logprobs": null,
"text": "Blue.\n\nThe completed quote is a well-known poetic phrase often used as a tongue-in-cheek romantic gesture. However, it's worth noting that true violets are actually purple in color, not blue. This phrase is a playful variation of the traditional \"Roses are red, violets are blue,\" which typically goes like this:\n\n\"Roses are red, violets are blue,\nSugar is sweet, and so are you.\"\n\nThis original quote has been used for centuries to make a lighthearted, whimsical compliment in poetry, songs, and spoken words." "text": "Blue.\n\nMy response is a play on words of the classic nursery rhyme \"Roses are red, violets are blue.\" In traditional rhymes, violets are typically depicted as being purple or blue in color. I've taken this convention and completed the sentence with the word \"blue\" to create a punny adaptation of the original phrase."
} }
], ],
"created": 1753762608, "created": 1753810618,
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"object": "text_completion", "object": "text_completion",
"system_fingerprint": "fp_ollama", "system_fingerprint": "fp_ollama",
"usage": { "usage": {
"completion_tokens": 120, "completion_tokens": 72,
"prompt_tokens": 50, "prompt_tokens": 50,
"total_tokens": 170, "total_tokens": 122,
"completion_tokens_details": null, "completion_tokens_details": null,
"prompt_tokens_details": null "prompt_tokens_details": null
} }
}
}, },
"is_streaming": false "is_streaming": false
} }

223
tests/integration/inference/recordings/responses/a6810c23eda8.json
View file

@ -20,8 +20,10 @@
"response": { "response": {
"body": [ "body": [
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.717175Z", "created_at": "2025-07-29T17:37:01.808564Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -33,10 +35,13 @@
"response": " ______", "response": " ______",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.759811Z", "created_at": "2025-07-29T17:37:01.850237Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -48,10 +53,13 @@
"response": "_", "response": "_",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.802135Z", "created_at": "2025-07-29T17:37:01.891462Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -63,10 +71,13 @@
"response": ".\n\n", "response": ".\n\n",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.843818Z", "created_at": "2025-07-29T17:37:01.9326Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -78,10 +89,13 @@
"response": "The", "response": "The",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.8848Z", "created_at": "2025-07-29T17:37:01.97369Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -93,10 +107,13 @@
"response": " best", "response": " best",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.926824Z", "created_at": "2025-07-29T17:37:02.014668Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -108,10 +125,13 @@
"response": " answer", "response": " answer",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:53.96764Z", "created_at": "2025-07-29T17:37:02.055932Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -123,10 +143,13 @@
"response": " is", "response": " is",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.008868Z", "created_at": "2025-07-29T17:37:02.097128Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -138,10 +161,13 @@
"response": " blue", "response": " blue",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.049584Z", "created_at": "2025-07-29T17:37:02.137979Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -153,10 +179,13 @@
"response": ".", "response": ".",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.090467Z", "created_at": "2025-07-29T17:37:02.178748Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -168,10 +197,13 @@
"response": " The", "response": " The",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.131216Z", "created_at": "2025-07-29T17:37:02.219748Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -183,10 +215,13 @@
"response": " traditional", "response": " traditional",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.171811Z", "created_at": "2025-07-29T17:37:02.260654Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -198,10 +233,13 @@
"response": " nursery", "response": " nursery",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.212449Z", "created_at": "2025-07-29T17:37:02.301472Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -213,10 +251,13 @@
"response": " rhyme", "response": " rhyme",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.254055Z", "created_at": "2025-07-29T17:37:02.342126Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -228,10 +269,13 @@
"response": " goes", "response": " goes",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.296182Z", "created_at": "2025-07-29T17:37:02.383051Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -243,10 +287,13 @@
"response": " like", "response": " like",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.339421Z", "created_at": "2025-07-29T17:37:02.423897Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -258,10 +305,13 @@
"response": " this", "response": " this",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.380632Z", "created_at": "2025-07-29T17:37:02.46492Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -273,10 +323,13 @@
"response": ":\n\n", "response": ":\n\n",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.423681Z", "created_at": "2025-07-29T17:37:02.505397Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -288,10 +341,13 @@
"response": "R", "response": "R",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.466032Z", "created_at": "2025-07-29T17:37:02.546389Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -303,10 +359,13 @@
"response": "oses", "response": "oses",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.508317Z", "created_at": "2025-07-29T17:37:02.587385Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -318,10 +377,13 @@
"response": " are", "response": " are",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.551009Z", "created_at": "2025-07-29T17:37:02.628982Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -333,10 +395,13 @@
"response": " red", "response": " red",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.595853Z", "created_at": "2025-07-29T17:37:02.670146Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -348,10 +413,13 @@
"response": ",\n", "response": ",\n",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.638044Z", "created_at": "2025-07-29T17:37:02.711586Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -363,10 +431,13 @@
"response": "V", "response": "V",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.679396Z", "created_at": "2025-07-29T17:37:02.753401Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -378,10 +449,13 @@
"response": "io", "response": "io",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.72096Z", "created_at": "2025-07-29T17:37:02.795085Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -393,10 +467,13 @@
"response": "lets", "response": "lets",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.763977Z", "created_at": "2025-07-29T17:37:02.836212Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -408,10 +485,13 @@
"response": " are", "response": " are",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.80598Z", "created_at": "2025-07-29T17:37:02.877229Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -423,10 +503,13 @@
"response": " blue", "response": " blue",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.847977Z", "created_at": "2025-07-29T17:37:02.919071Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -438,10 +521,13 @@
"response": ",\n", "response": ",\n",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.890637Z", "created_at": "2025-07-29T17:37:02.962816Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -453,10 +539,13 @@
"response": "Sugar", "response": "Sugar",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.931597Z", "created_at": "2025-07-29T17:37:03.004934Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -468,10 +557,13 @@
"response": " is", "response": " is",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:54.972266Z", "created_at": "2025-07-29T17:37:03.047274Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -483,10 +575,13 @@
"response": " sweet", "response": " sweet",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.01467Z", "created_at": "2025-07-29T17:37:03.088469Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -498,10 +593,13 @@
"response": ",\n", "response": ",\n",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.055561Z", "created_at": "2025-07-29T17:37:03.129445Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -513,10 +611,13 @@
"response": "And", "response": "And",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.097823Z", "created_at": "2025-07-29T17:37:03.170337Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -528,10 +629,13 @@
"response": " so", "response": " so",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.139389Z", "created_at": "2025-07-29T17:37:03.214189Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -543,10 +647,13 @@
"response": " are", "response": " are",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.181536Z", "created_at": "2025-07-29T17:37:03.259789Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -558,10 +665,13 @@
"response": " you", "response": " you",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.224644Z", "created_at": "2025-07-29T17:37:03.311454Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -573,10 +683,13 @@
"response": "!", "response": "!",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.267976Z", "created_at": "2025-07-29T17:37:03.358855Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -588,10 +701,13 @@
"response": " (", "response": " (",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.311629Z", "created_at": "2025-07-29T17:37:03.405774Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -603,10 +719,13 @@
"response": "Or", "response": "Or",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.355343Z", "created_at": "2025-07-29T17:37:03.449402Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -618,10 +737,13 @@
"response": " something", "response": " something",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.396541Z", "created_at": "2025-07-29T17:37:03.493595Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -633,10 +755,13 @@
"response": " similar", "response": " similar",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.437899Z", "created_at": "2025-07-29T17:37:03.536236Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -648,22 +773,26 @@
"response": ".)", "response": ".)",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:16:55.479038Z", "created_at": "2025-07-29T17:37:03.578174Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 1881416167, "total_duration": 1877723375,
"load_duration": 69754000, "load_duration": 64346709,
"prompt_eval_count": 18, "prompt_eval_count": 18,
"prompt_eval_duration": 46361125, "prompt_eval_duration": 41320958,
"eval_count": 43, "eval_count": 43,
"eval_duration": 1762259458, "eval_duration": 1771501375,
"response": "", "response": "",
"thinking": null, "thinking": null,
"context": null "context": null
} }
}
], ],
"is_streaming": true "is_streaming": true
} }

13
tests/integration/inference/recordings/responses/ae6835cfe70e.json
View file

@ -17,19 +17,22 @@
}, },
"response": { "response": {
"body": { "body": {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:22.73932Z", "created_at": "2025-07-29T17:37:31.563015Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 660872000, "total_duration": 697289250,
"load_duration": 76282083, "load_duration": 84472834,
"prompt_eval_count": 386, "prompt_eval_count": 386,
"prompt_eval_duration": 541896167, "prompt_eval_duration": 564623708,
"eval_count": 2, "eval_count": 2,
"eval_duration": 42127791, "eval_duration": 47500459,
"response": "[]", "response": "[]",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
"is_streaming": false "is_streaming": false
} }

63
tests/integration/inference/recordings/responses/b91f1fb4aedb.json
View file

@ -18,8 +18,10 @@
"response": { "response": {
"body": [ "body": [
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.217546Z", "created_at": "2025-07-29T17:37:08.301403Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -31,10 +33,13 @@
"response": "[", "response": "[",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.267879Z", "created_at": "2025-07-29T17:37:08.344225Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -46,10 +51,13 @@
"response": "get", "response": "get",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.315525Z", "created_at": "2025-07-29T17:37:08.38649Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -61,10 +69,13 @@
"response": "_weather", "response": "_weather",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.362669Z", "created_at": "2025-07-29T17:37:08.42879Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -76,10 +87,13 @@
"response": "(location", "response": "(location",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.406139Z", "created_at": "2025-07-29T17:37:08.470562Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -91,10 +105,13 @@
"response": "=\"", "response": "=\"",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.450302Z", "created_at": "2025-07-29T17:37:08.512144Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -106,10 +123,13 @@
"response": "San", "response": "San",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.496893Z", "created_at": "2025-07-29T17:37:08.553706Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -121,10 +141,13 @@
"response": " Francisco", "response": " Francisco",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.540977Z", "created_at": "2025-07-29T17:37:08.59536Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -136,10 +159,13 @@
"response": ",", "response": ",",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.586272Z", "created_at": "2025-07-29T17:37:08.636886Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -151,10 +177,13 @@
"response": " CA", "response": " CA",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.631743Z", "created_at": "2025-07-29T17:37:08.678935Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -166,22 +195,26 @@
"response": "\")]", "response": "\")]",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:00.676251Z", "created_at": "2025-07-29T17:37:08.719919Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 682827167, "total_duration": 609334000,
"load_duration": 111852875, "load_duration": 85744542,
"prompt_eval_count": 339, "prompt_eval_count": 339,
"prompt_eval_duration": 109521833, "prompt_eval_duration": 102984708,
"eval_count": 11, "eval_count": 11,
"eval_duration": 460495042, "eval_duration": 420012834,
"response": "", "response": "",
"thinking": null, "thinking": null,
"context": null "context": null
} }
}
], ],
"is_streaming": true "is_streaming": true
} }

1153
tests/integration/inference/recordings/responses/bbd0637dce16.json
View file

File diff suppressed because it is too large Load diff

18
tests/integration/inference/recordings/responses/dd9e7d5913e9.json
View file

@ -18,8 +18,10 @@
"response": { "response": {
"body": [ "body": [
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:14.122273Z", "created_at": "2025-07-29T17:37:22.214492Z",
"done": false, "done": false,
"done_reason": null, "done_reason": null,
"total_duration": null, "total_duration": null,
@ -31,22 +33,26 @@
"response": "[]", "response": "[]",
"thinking": null, "thinking": null,
"context": null "context": null
}
}, },
{ {
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16", "model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-29T04:17:14.165968Z", "created_at": "2025-07-29T17:37:22.25685Z",
"done": true, "done": true,
"done_reason": "stop", "done_reason": "stop",
"total_duration": 663520959, "total_duration": 690090375,
"load_duration": 67474917, "load_duration": 101933500,
"prompt_eval_count": 386, "prompt_eval_count": 386,
"prompt_eval_duration": 545132042, "prompt_eval_duration": 544455708,
"eval_count": 2, "eval_count": 2,
"eval_duration": 50234083, "eval_duration": 42919375,
"response": "", "response": "",
"thinking": null, "thinking": null,
"context": null "context": null
} }
}
], ],
"is_streaming": true "is_streaming": true
} }

93
tests/integration/test_inference_recordings.py
View file

@ -7,11 +7,20 @@
import sqlite3 import sqlite3
import tempfile import tempfile
from pathlib import Path from pathlib import Path
from unittest.mock import Mock, patch from unittest.mock import patch
import pytest import pytest
from openai import AsyncOpenAI from openai import AsyncOpenAI
# Import the real Pydantic response types instead of using Mocks
from llama_stack.apis.inference import (
OpenAIAssistantMessageParam,
OpenAIChatCompletion,
OpenAIChoice,
OpenAIEmbeddingData,
OpenAIEmbeddingsResponse,
OpenAIEmbeddingUsage,
)
from llama_stack.testing.inference_recorder import ( from llama_stack.testing.inference_recorder import (
ResponseStorage, ResponseStorage,
inference_recording, inference_recording,
@ -27,44 +36,36 @@ def temp_storage_dir():
@pytest.fixture @pytest.fixture
def mock_openai_response(): def real_openai_chat_response():
"""Mock OpenAI response object.""" """Real OpenAI chat completion response using proper Pydantic objects."""
mock_response = Mock() return OpenAIChatCompletion(
mock_response.choices = [Mock()] id="chatcmpl-test123",
mock_response.choices[0].message.content = "Hello! I'm doing well, thank you for asking." choices=[
mock_response.model_dump.return_value = { OpenAIChoice(
"id": "chatcmpl-test123", index=0,
"object": "chat.completion", message=OpenAIAssistantMessageParam(
"choices": [ role="assistant", content="Hello! I'm doing well, thank you for asking."
{ ),
"index": 0, finish_reason="stop",
"message": {"role": "assistant", "content": "Hello! I'm doing well, thank you for asking."}, )
"finish_reason": "stop",
}
], ],
"model": "llama3.2:3b", created=1234567890,
"usage": {"prompt_tokens": 10, "completion_tokens": 15, "total_tokens": 25}, model="llama3.2:3b",
} )
return mock_response
@pytest.fixture @pytest.fixture
def mock_embeddings_response(): def real_embeddings_response():
"""Mock OpenAI embeddings response object.""" """Real OpenAI embeddings response using proper Pydantic objects."""
mock_response = Mock() return OpenAIEmbeddingsResponse(
mock_response.data = [Mock(embedding=[0.1, 0.2, 0.3]), Mock(embedding=[0.4, 0.5, 0.6])] object="list",
mock_response.model_dump.return_value = { data=[
"object": "list", OpenAIEmbeddingData(object="embedding", embedding=[0.1, 0.2, 0.3], index=0),
"data": [ OpenAIEmbeddingData(object="embedding", embedding=[0.4, 0.5, 0.6], index=1),
{"object": "embedding", "embedding": [0.1, 0.2, 0.3], "index": 0},
{"object": "embedding", "embedding": [0.4, 0.5, 0.6], "index": 1},
], ],
"model": "nomic-embed-text", model="nomic-embed-text",
"usage": {"prompt_tokens": 6, "total_tokens": 6}, usage=OpenAIEmbeddingUsage(prompt_tokens=6, total_tokens=6),
} )
return mock_response
class TestInferenceRecording: class TestInferenceRecording:
@ -160,11 +161,11 @@ class TestInferenceRecording:
assert retrieved["request"]["model"] == "llama3.2:3b" assert retrieved["request"]["model"] == "llama3.2:3b"
assert retrieved["response"]["body"]["content"] == "test response" assert retrieved["response"]["body"]["content"] == "test response"
async def test_recording_mode(self, temp_storage_dir, mock_openai_response): async def test_recording_mode(self, temp_storage_dir, real_openai_chat_response):
"""Test that recording mode captures and stores responses.""" """Test that recording mode captures and stores responses."""
async def mock_create(*args, **kwargs): async def mock_create(*args, **kwargs):
return mock_openai_response return real_openai_chat_response
temp_storage_dir = temp_storage_dir / "test_recording_mode" temp_storage_dir = temp_storage_dir / "test_recording_mode"
with patch("openai.resources.chat.completions.AsyncCompletions.create", side_effect=mock_create): with patch("openai.resources.chat.completions.AsyncCompletions.create", side_effect=mock_create):
@ -188,11 +189,11 @@ class TestInferenceRecording:
assert recordings == 1 assert recordings == 1
async def test_replay_mode(self, temp_storage_dir, mock_openai_response): async def test_replay_mode(self, temp_storage_dir, real_openai_chat_response):
"""Test that replay mode returns stored responses without making real calls.""" """Test that replay mode returns stored responses without making real calls."""
async def mock_create(*args, **kwargs): async def mock_create(*args, **kwargs):
return mock_openai_response return real_openai_chat_response
temp_storage_dir = temp_storage_dir / "test_replay_mode" temp_storage_dir = temp_storage_dir / "test_replay_mode"
# First, record a response # First, record a response
@ -200,7 +201,7 @@ class TestInferenceRecording:
with inference_recording(mode="record", storage_dir=str(temp_storage_dir)): with inference_recording(mode="record", storage_dir=str(temp_storage_dir)):
client = AsyncOpenAI(base_url="http://localhost:11434/v1", api_key="test") client = AsyncOpenAI(base_url="http://localhost:11434/v1", api_key="test")
await client.chat.completions.create( response = await client.chat.completions.create(
model="llama3.2:3b", model="llama3.2:3b",
messages=[{"role": "user", "content": "Hello, how are you?"}], messages=[{"role": "user", "content": "Hello, how are you?"}],
temperature=0.7, temperature=0.7,
@ -220,7 +221,7 @@ class TestInferenceRecording:
) )
# Verify we got the recorded response # Verify we got the recorded response
assert response["choices"][0]["message"]["content"] == "Hello! I'm doing well, thank you for asking." assert response.choices[0].message.content == "Hello! I'm doing well, thank you for asking."
# Verify the original method was NOT called # Verify the original method was NOT called
mock_create_patch.assert_not_called() mock_create_patch.assert_not_called()
@ -237,11 +238,11 @@ class TestInferenceRecording:
model="llama3.2:3b", messages=[{"role": "user", "content": "This was never recorded"}] model="llama3.2:3b", messages=[{"role": "user", "content": "This was never recorded"}]
) )
async def test_embeddings_recording(self, temp_storage_dir, mock_embeddings_response): async def test_embeddings_recording(self, temp_storage_dir, real_embeddings_response):
"""Test recording and replay of embeddings calls.""" """Test recording and replay of embeddings calls."""
async def mock_create(*args, **kwargs): async def mock_create(*args, **kwargs):
return mock_embeddings_response return real_embeddings_response
temp_storage_dir = temp_storage_dir / "test_embeddings_recording" temp_storage_dir = temp_storage_dir / "test_embeddings_recording"
# Record # Record
@ -265,17 +266,17 @@ class TestInferenceRecording:
) )
# Verify we got the recorded response # Verify we got the recorded response
assert len(response["data"]) == 2 assert len(response.data) == 2
assert response["data"][0]["embedding"] == [0.1, 0.2, 0.3] assert response.data[0].embedding == [0.1, 0.2, 0.3]
# Verify original method was not called # Verify original method was not called
mock_create_patch.assert_not_called() mock_create_patch.assert_not_called()
async def test_live_mode(self, mock_openai_response): async def test_live_mode(self, real_openai_chat_response):
"""Test that live mode passes through to original methods.""" """Test that live mode passes through to original methods."""
async def mock_create(*args, **kwargs): async def mock_create(*args, **kwargs):
return mock_openai_response return real_openai_chat_response
with patch("openai.resources.chat.completions.AsyncCompletions.create", side_effect=mock_create): with patch("openai.resources.chat.completions.AsyncCompletions.create", side_effect=mock_create):
with inference_recording(mode="live"): with inference_recording(mode="live"):