Remove deprecated OpenAILLMContext as well as everything (code paths or whole types) dependent on it (all of which were also deprecated)
This commit is contained in:
Paul Kompfner
File diff suppressed because it is too large
Load Diff
@@ -4,13 +4,16 @@
|
||||
# SPDX-License-Identifier: BSD 2-Clause License
|
||||
#
|
||||
|
||||
import json
|
||||
import unittest
|
||||
|
||||
from pipecat.frames.frames import (
|
||||
BotStartedSpeakingFrame,
|
||||
BotStoppedSpeakingFrame,
|
||||
FunctionCallFromLLM,
|
||||
FunctionCallInProgressFrame,
|
||||
FunctionCallResultFrame,
|
||||
FunctionCallResultProperties,
|
||||
FunctionCallsStartedFrame,
|
||||
InterimTranscriptionFrame,
|
||||
InterruptionFrame,
|
||||
@@ -26,6 +29,7 @@ from pipecat.frames.frames import (
|
||||
LLMThoughtStartFrame,
|
||||
LLMThoughtTextFrame,
|
||||
StartFrame,
|
||||
TextFrame,
|
||||
TranscriptionFrame,
|
||||
TranslationFrame,
|
||||
UserMuteStartedFrame,
|
||||
@@ -588,6 +592,165 @@ class TestLLMAssistantAggregator(unittest.IsolatedAsyncioTestCase):
|
||||
self.assertTrue(should_stop)
|
||||
self.assertEqual(stop_message.content, "Hello from Pipecat!")
|
||||
|
||||
async def test_multiple_text_with_spaces(self):
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
|
||||
def make_text_frame(text: str) -> TextFrame:
|
||||
frame = TextFrame(text=text)
|
||||
frame.includes_inter_frame_spaces = True
|
||||
return frame
|
||||
|
||||
frames_to_send = [
|
||||
LLMFullResponseStartFrame(),
|
||||
make_text_frame("Hello "),
|
||||
make_text_frame("Pipecat. "),
|
||||
make_text_frame("How are "),
|
||||
make_text_frame("you?"),
|
||||
LLMFullResponseEndFrame(),
|
||||
]
|
||||
expected_down_frames = [LLMContextFrame, LLMContextAssistantTimestampFrame]
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert context.messages[0]["content"] == "Hello Pipecat. How are you?"
|
||||
|
||||
async def test_multiple_text_stripped(self):
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
frames_to_send = [
|
||||
LLMFullResponseStartFrame(),
|
||||
TextFrame(text="Hello"),
|
||||
TextFrame(text="Pipecat."),
|
||||
TextFrame(text="How are"),
|
||||
TextFrame(text="you?"),
|
||||
LLMFullResponseEndFrame(),
|
||||
]
|
||||
expected_down_frames = [LLMContextFrame, LLMContextAssistantTimestampFrame]
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert context.messages[0]["content"] == "Hello Pipecat. How are you?"
|
||||
|
||||
async def test_multiple_text_mixed_spaces(self):
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
|
||||
def make_text_frame(text: str, includes_spaces: bool) -> TextFrame:
|
||||
frame = TextFrame(text=text)
|
||||
frame.includes_inter_frame_spaces = includes_spaces
|
||||
return frame
|
||||
|
||||
frames_to_send = [
|
||||
LLMFullResponseStartFrame(),
|
||||
make_text_frame("Hello ", includes_spaces=True),
|
||||
make_text_frame("Pipecat. ", includes_spaces=True),
|
||||
make_text_frame("Here's some", includes_spaces=True),
|
||||
make_text_frame(
|
||||
" code:", includes_spaces=True
|
||||
), # Validates ending includes_inter_frame_spaces run with no space
|
||||
make_text_frame("```python\nprint('Hello, World!')\n```", includes_spaces=False),
|
||||
make_text_frame(
|
||||
"```javascript\nconsole.log('Hello, World!');\n```", includes_spaces=False
|
||||
),
|
||||
make_text_frame(
|
||||
" And some more: ", includes_spaces=True
|
||||
), # Validates starting includes_inter_frame_spaces run with a space and ending it with no space
|
||||
make_text_frame("```html\n<div>Hello, World!</div>\n```", includes_spaces=False),
|
||||
make_text_frame(
|
||||
"Hope that ", includes_spaces=True
|
||||
), # Validates starting includes_inter_frame_spaces run with no space
|
||||
make_text_frame("helps!", includes_spaces=True),
|
||||
LLMFullResponseEndFrame(),
|
||||
]
|
||||
expected_down_frames = [LLMContextFrame, LLMContextAssistantTimestampFrame]
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert context.messages[0]["content"] == (
|
||||
"Hello Pipecat. Here's some code: "
|
||||
"```python\nprint('Hello, World!')\n``` "
|
||||
"```javascript\nconsole.log('Hello, World!');\n``` "
|
||||
"And some more: "
|
||||
"```html\n<div>Hello, World!</div>\n``` "
|
||||
"Hope that helps!"
|
||||
)
|
||||
|
||||
async def test_multiple_responses(self):
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
|
||||
def make_text_frame(text: str) -> TextFrame:
|
||||
frame = TextFrame(text=text)
|
||||
frame.includes_inter_frame_spaces = True
|
||||
return frame
|
||||
|
||||
frames_to_send = [
|
||||
LLMFullResponseStartFrame(),
|
||||
make_text_frame("Hello "),
|
||||
make_text_frame("Pipecat."),
|
||||
LLMFullResponseEndFrame(),
|
||||
LLMFullResponseStartFrame(),
|
||||
make_text_frame(text="How are "),
|
||||
make_text_frame(text="you?"),
|
||||
LLMFullResponseEndFrame(),
|
||||
]
|
||||
expected_down_frames = [
|
||||
LLMContextFrame,
|
||||
LLMContextAssistantTimestampFrame,
|
||||
LLMContextFrame,
|
||||
LLMContextAssistantTimestampFrame,
|
||||
]
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert context.messages[0]["content"] == "Hello Pipecat."
|
||||
assert context.messages[1]["content"] == "How are you?"
|
||||
|
||||
async def test_multiple_responses_interruption(self):
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
|
||||
def make_text_frame(text: str) -> TextFrame:
|
||||
frame = TextFrame(text=text)
|
||||
frame.includes_inter_frame_spaces = True
|
||||
return frame
|
||||
|
||||
frames_to_send = [
|
||||
LLMFullResponseStartFrame(),
|
||||
make_text_frame("Hello "),
|
||||
make_text_frame("Pipecat."),
|
||||
LLMFullResponseEndFrame(),
|
||||
SleepFrame(0.15),
|
||||
InterruptionFrame(),
|
||||
LLMFullResponseStartFrame(),
|
||||
make_text_frame("How are "),
|
||||
make_text_frame("you?"),
|
||||
LLMFullResponseEndFrame(),
|
||||
]
|
||||
expected_down_frames = [
|
||||
LLMContextFrame,
|
||||
LLMContextAssistantTimestampFrame,
|
||||
InterruptionFrame,
|
||||
LLMContextFrame,
|
||||
LLMContextAssistantTimestampFrame,
|
||||
]
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert context.messages[0]["content"] == "Hello Pipecat."
|
||||
assert context.messages[1]["content"] == "How are you?"
|
||||
|
||||
async def test_interruption(self):
|
||||
context = LLMContext()
|
||||
|
||||
@@ -635,6 +798,67 @@ class TestLLMAssistantAggregator(unittest.IsolatedAsyncioTestCase):
|
||||
self.assertEqual(stop_messages[0].content, "Hello")
|
||||
self.assertEqual(stop_messages[1].content, "Hello there!")
|
||||
|
||||
async def test_function_call(self):
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
frames_to_send = [
|
||||
FunctionCallInProgressFrame(
|
||||
function_name="get_weather",
|
||||
tool_call_id="1",
|
||||
arguments={"location": "Los Angeles"},
|
||||
cancel_on_interruption=False,
|
||||
),
|
||||
SleepFrame(),
|
||||
FunctionCallResultFrame(
|
||||
function_name="get_weather",
|
||||
tool_call_id="1",
|
||||
arguments={"location": "Los Angeles"},
|
||||
result={"conditions": "Sunny"},
|
||||
),
|
||||
]
|
||||
expected_down_frames = []
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert json.loads(context.messages[-1]["content"]) == {"conditions": "Sunny"}
|
||||
|
||||
async def test_function_call_on_context_updated(self):
|
||||
context_updated = False
|
||||
|
||||
async def on_context_updated():
|
||||
nonlocal context_updated
|
||||
context_updated = True
|
||||
|
||||
context = LLMContext()
|
||||
aggregator = LLMAssistantAggregator(context)
|
||||
frames_to_send = [
|
||||
FunctionCallInProgressFrame(
|
||||
function_name="get_weather",
|
||||
tool_call_id="1",
|
||||
arguments={"location": "Los Angeles"},
|
||||
cancel_on_interruption=False,
|
||||
),
|
||||
SleepFrame(),
|
||||
FunctionCallResultFrame(
|
||||
function_name="get_weather",
|
||||
tool_call_id="1",
|
||||
arguments={"location": "Los Angeles"},
|
||||
result={"conditions": "Sunny"},
|
||||
properties=FunctionCallResultProperties(on_context_updated=on_context_updated),
|
||||
),
|
||||
SleepFrame(),
|
||||
]
|
||||
expected_down_frames = []
|
||||
await run_test(
|
||||
aggregator,
|
||||
frames_to_send=frames_to_send,
|
||||
expected_down_frames=expected_down_frames,
|
||||
)
|
||||
assert json.loads(context.messages[-1]["content"]) == {"conditions": "Sunny"}
|
||||
assert context_updated
|
||||
|
||||
async def test_thought(self):
|
||||
context = LLMContext()
|
||||
|
||||
|
||||
@@ -1,81 +0,0 @@
|
||||
#
|
||||
# Copyright (c) 2024-2026, Daily
|
||||
#
|
||||
# SPDX-License-Identifier: BSD 2-Clause License
|
||||
#
|
||||
|
||||
"""Unit tests for Google LLM OpenAI Beta service."""
|
||||
|
||||
import asyncio
|
||||
import warnings
|
||||
from unittest.mock import AsyncMock, patch
|
||||
|
||||
import pytest
|
||||
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
|
||||
try:
|
||||
from pipecat.services.google.openai.llm import GoogleLLMOpenAIBetaService
|
||||
|
||||
google_available = True
|
||||
except Exception:
|
||||
google_available = False
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
@pytest.mark.skipif(not google_available, reason="Google dependencies not installed")
|
||||
async def test_google_llm_openai_stream_closed_on_cancellation():
|
||||
"""Test that the stream is closed when CancelledError occurs during iteration.
|
||||
|
||||
This prevents socket leaks when the pipeline is interrupted (e.g., user interruption).
|
||||
See issue #3639.
|
||||
"""
|
||||
with patch.object(GoogleLLMOpenAIBetaService, "create_client"):
|
||||
with warnings.catch_warnings():
|
||||
warnings.simplefilter("ignore", DeprecationWarning)
|
||||
service = GoogleLLMOpenAIBetaService(api_key="test-key", model="test-model")
|
||||
service._client = AsyncMock()
|
||||
|
||||
stream_closed = False
|
||||
|
||||
class MockAsyncStream:
|
||||
"""Mock AsyncStream that tracks close() calls and raises CancelledError."""
|
||||
|
||||
def __init__(self):
|
||||
self.iteration_count = 0
|
||||
|
||||
async def __aenter__(self):
|
||||
return self
|
||||
|
||||
async def __aexit__(self, exc_type, exc_val, exc_tb):
|
||||
nonlocal stream_closed
|
||||
stream_closed = True
|
||||
return False
|
||||
|
||||
def __aiter__(self):
|
||||
return self
|
||||
|
||||
async def __anext__(self):
|
||||
self.iteration_count += 1
|
||||
if self.iteration_count > 1:
|
||||
raise asyncio.CancelledError()
|
||||
mock_chunk = AsyncMock()
|
||||
mock_chunk.usage = None
|
||||
mock_chunk.choices = []
|
||||
return mock_chunk
|
||||
|
||||
mock_stream = MockAsyncStream()
|
||||
|
||||
service._stream_chat_completions_specific_context = AsyncMock(return_value=mock_stream)
|
||||
service.start_ttfb_metrics = AsyncMock()
|
||||
service.stop_ttfb_metrics = AsyncMock()
|
||||
service.start_llm_usage_metrics = AsyncMock()
|
||||
|
||||
context = OpenAILLMContext(
|
||||
messages=[{"role": "user", "content": "Hello"}],
|
||||
)
|
||||
|
||||
with pytest.raises(asyncio.CancelledError):
|
||||
await service._process_context(context)
|
||||
|
||||
assert stream_closed, "Stream should be closed even when CancelledError occurs"
|
||||
@@ -84,61 +84,6 @@ async def test_openai_run_inference_with_llm_context():
|
||||
)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_openai_run_inference_with_openai_llm_context():
|
||||
"""Test run_inference with OpenAILLMContext returns expected response."""
|
||||
# Create service with mocked client and specific parameters
|
||||
with patch.object(OpenAILLMService, "create_client"):
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
from pipecat.services.openai.base_llm import BaseOpenAILLMService
|
||||
|
||||
params = BaseOpenAILLMService.InputParams(
|
||||
temperature=0.8, max_completion_tokens=150, presence_penalty=0.3, top_p=0.9
|
||||
)
|
||||
service = OpenAILLMService(model="gpt-4", params=params)
|
||||
service._client = AsyncMock()
|
||||
|
||||
# Create OpenAILLMContext
|
||||
context = OpenAILLMContext(
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a helpful assistant"},
|
||||
{"role": "user", "content": "Hello, world!"},
|
||||
],
|
||||
tools=OPENAI_NOT_GIVEN,
|
||||
tool_choice=OPENAI_NOT_GIVEN,
|
||||
)
|
||||
|
||||
# Mock response
|
||||
mock_response = MagicMock()
|
||||
mock_response.choices = [MagicMock()]
|
||||
mock_response.choices[0].message.content = "Hello! How can I help you today?"
|
||||
service._client.chat.completions.create.return_value = mock_response
|
||||
|
||||
# Execute
|
||||
result = await service.run_inference(context)
|
||||
|
||||
# Verify
|
||||
assert result == "Hello! How can I help you today?"
|
||||
service._client.chat.completions.create.assert_called_once_with(
|
||||
model="gpt-4",
|
||||
stream=False,
|
||||
frequency_penalty=OPENAI_NOT_GIVEN,
|
||||
presence_penalty=0.3,
|
||||
seed=OPENAI_NOT_GIVEN,
|
||||
temperature=0.8,
|
||||
top_p=0.9,
|
||||
max_tokens=OPENAI_NOT_GIVEN,
|
||||
max_completion_tokens=150,
|
||||
service_tier=OPENAI_NOT_GIVEN,
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a helpful assistant"},
|
||||
{"role": "user", "content": "Hello, world!"},
|
||||
],
|
||||
tools=OPENAI_NOT_GIVEN,
|
||||
tool_choice=OPENAI_NOT_GIVEN,
|
||||
)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_openai_run_inference_client_exception():
|
||||
"""Test that exceptions from the client are propagated."""
|
||||
@@ -209,54 +154,6 @@ async def test_anthropic_run_inference_with_llm_context():
|
||||
)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_anthropic_run_inference_with_openai_llm_context():
|
||||
"""Test run_inference with OpenAILLMContext returns expected response for Anthropic."""
|
||||
# Create service with mocked client and specific parameters
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
from pipecat.services.anthropic.llm import AnthropicLLMService
|
||||
|
||||
params = AnthropicLLMService.InputParams(max_tokens=1024, temperature=0.7, top_k=40, top_p=0.9)
|
||||
service = AnthropicLLMService(
|
||||
api_key="test-key", model="claude-3-sonnet-20240229", params=params
|
||||
)
|
||||
service._client = AsyncMock()
|
||||
|
||||
# Create OpenAILLMContext
|
||||
context = OpenAILLMContext(
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a helpful assistant"},
|
||||
{"role": "user", "content": "Hello, world!"},
|
||||
],
|
||||
tools=NOT_GIVEN,
|
||||
tool_choice=NOT_GIVEN,
|
||||
)
|
||||
|
||||
# Mock response
|
||||
mock_response = MagicMock()
|
||||
mock_response.content = [MagicMock()]
|
||||
mock_response.content[0].text = "Hello! How can I help you today?"
|
||||
service._client.beta.messages.create.return_value = mock_response
|
||||
|
||||
# Execute
|
||||
result = await service.run_inference(context)
|
||||
|
||||
# Verify
|
||||
assert result == "Hello! How can I help you today?"
|
||||
service._client.beta.messages.create.assert_called_once_with(
|
||||
model="claude-3-sonnet-20240229",
|
||||
max_tokens=1024,
|
||||
stream=False,
|
||||
temperature=0.7,
|
||||
top_k=40,
|
||||
top_p=0.9,
|
||||
messages=[{"role": "user", "content": "Hello, world!"}],
|
||||
system="You are a helpful assistant",
|
||||
tools=[],
|
||||
betas=["interleaved-thinking-2025-05-14"],
|
||||
)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_anthropic_run_inference_client_exception():
|
||||
"""Test that exceptions from the Anthropic client are propagated."""
|
||||
@@ -336,61 +233,6 @@ async def test_google_run_inference_client_exception():
|
||||
await service.run_inference(mock_context)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_google_run_inference_with_openai_llm_context():
|
||||
"""Test run_inference with OpenAILLMContext returns expected response for Google."""
|
||||
# Create service with mocked client and specific parameters
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
|
||||
params = GoogleLLMService.InputParams(max_tokens=256, temperature=0.4, top_k=30, top_p=0.75)
|
||||
service = GoogleLLMService(api_key="test-key", model="gemini-2.0-flash", params=params)
|
||||
service._client = AsyncMock()
|
||||
|
||||
# Create OpenAILLMContext
|
||||
context = OpenAILLMContext(
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a helpful assistant"},
|
||||
{"role": "user", "content": "Hello, world!"},
|
||||
],
|
||||
tools=NOT_GIVEN,
|
||||
tool_choice=NOT_GIVEN,
|
||||
)
|
||||
|
||||
# Mock response
|
||||
mock_response = MagicMock()
|
||||
mock_response.candidates = [MagicMock()]
|
||||
mock_response.candidates[0].content = MagicMock()
|
||||
mock_response.candidates[0].content.parts = [MagicMock()]
|
||||
mock_response.candidates[0].content.parts[0].text = "Hello! How can I help you today?"
|
||||
service._client.aio = AsyncMock()
|
||||
service._client.aio.models = AsyncMock()
|
||||
service._client.aio.models.generate_content = AsyncMock(return_value=mock_response)
|
||||
|
||||
# Execute
|
||||
result = await service.run_inference(context)
|
||||
|
||||
# Verify
|
||||
assert result == "Hello! How can I help you today?"
|
||||
|
||||
# Verify the call includes configured parameters
|
||||
call_kwargs = service._client.aio.models.generate_content.call_args.kwargs
|
||||
assert call_kwargs["model"] == "gemini-2.0-flash"
|
||||
# Contents is a Google Content object, so check its structure
|
||||
contents = call_kwargs["contents"]
|
||||
assert len(contents) == 1
|
||||
assert contents[0].role == "user"
|
||||
assert len(contents[0].parts) == 1
|
||||
assert contents[0].parts[0].text == "Hello, world!"
|
||||
assert "config" in call_kwargs
|
||||
config = call_kwargs["config"]
|
||||
# Config is a GenerateContentConfig object, so access attributes
|
||||
assert config.system_instruction == "You are a helpful assistant"
|
||||
assert config.temperature == 0.4
|
||||
assert config.top_k == 30
|
||||
assert config.top_p == 0.75
|
||||
assert config.max_output_tokens == 256
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_aws_bedrock_run_inference_with_llm_context():
|
||||
"""Test run_inference with LLMContext returns expected response for AWS Bedrock."""
|
||||
@@ -445,57 +287,6 @@ async def test_aws_bedrock_run_inference_with_llm_context():
|
||||
assert call_kwargs["inferenceConfig"]["topP"] == 0.85
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_aws_bedrock_run_inference_with_openai_llm_context():
|
||||
"""Test run_inference with OpenAILLMContext returns expected response for AWS Bedrock."""
|
||||
# Create service with specific parameters
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
from pipecat.services.aws.llm import AWSBedrockLLMService
|
||||
|
||||
params = AWSBedrockLLMService.InputParams(max_tokens=512, temperature=0.8, top_p=0.95)
|
||||
service = AWSBedrockLLMService(model="anthropic.claude-3-sonnet-20240229-v1:0", params=params)
|
||||
|
||||
# Create OpenAILLMContext
|
||||
context = OpenAILLMContext(
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a helpful assistant"},
|
||||
{"role": "user", "content": "Hello, world!"},
|
||||
],
|
||||
tools=NOT_GIVEN,
|
||||
tool_choice=NOT_GIVEN,
|
||||
)
|
||||
|
||||
# Mock the client and response
|
||||
mock_client = AsyncMock()
|
||||
mock_response = {
|
||||
"output": {"message": {"content": [{"text": "Hello! How can I help you today?"}]}}
|
||||
}
|
||||
mock_client.converse.return_value = mock_response
|
||||
|
||||
# Patch the _aws_session.client method to be an async context manager
|
||||
mock_context_manager = AsyncMock()
|
||||
mock_context_manager.__aenter__ = AsyncMock(return_value=mock_client)
|
||||
mock_context_manager.__aexit__ = AsyncMock(return_value=None)
|
||||
|
||||
with patch.object(service._aws_session, "client", return_value=mock_context_manager):
|
||||
# Execute
|
||||
result = await service.run_inference(context)
|
||||
|
||||
# Verify
|
||||
assert result == "Hello! How can I help you today?"
|
||||
|
||||
# Verify the call includes configured parameters
|
||||
call_kwargs = mock_client.converse.call_args.kwargs
|
||||
assert call_kwargs["modelId"] == "anthropic.claude-3-sonnet-20240229-v1:0"
|
||||
assert call_kwargs["messages"] == [{"role": "user", "content": [{"text": "Hello, world!"}]}]
|
||||
assert call_kwargs["system"] == [{"text": "You are a helpful assistant"}]
|
||||
assert call_kwargs["additionalModelRequestFields"] == {}
|
||||
assert "inferenceConfig" in call_kwargs
|
||||
assert call_kwargs["inferenceConfig"]["maxTokens"] == 512
|
||||
assert call_kwargs["inferenceConfig"]["temperature"] == 0.8
|
||||
assert call_kwargs["inferenceConfig"]["topP"] == 0.95
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_aws_bedrock_run_inference_client_exception():
|
||||
"""Test that exceptions from the AWS Bedrock client are propagated."""
|
||||
|
||||
Reference in New Issue
Block a user