Merge pull request #2408 from pipecat-ai/mb/add-mistral-llm
Add MistralLLMService
This commit is contained in:
Mark Backman
@@ -7,11 +7,19 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
|
||||
|
||||
## [Unreleased]
|
||||
|
||||
### Added
|
||||
|
||||
- Added `MistralLLMService`, using Mistral's chat completion API.
|
||||
|
||||
### Fixed
|
||||
|
||||
- Fixed an issue where `AsyncAITTSService` had very high latency in responding
|
||||
by adding `force=true` when sending the flush command.
|
||||
|
||||
### Other
|
||||
|
||||
- Added `14w-function-calling-mistal.py` using `MistralLLMService`.
|
||||
|
||||
## [0.0.80] - 2025-08-13
|
||||
|
||||
### Added
|
||||
|
||||
@@ -54,7 +54,7 @@ You can connect to Pipecat from any platform using our official SDKs:
|
||||
| Category | Services |
|
||||
| ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| Speech-to-Text | [AssemblyAI](https://docs.pipecat.ai/server/services/stt/assemblyai), [AWS](https://docs.pipecat.ai/server/services/stt/aws), [Azure](https://docs.pipecat.ai/server/services/stt/azure), [Cartesia](https://docs.pipecat.ai/server/services/stt/cartesia), [Deepgram](https://docs.pipecat.ai/server/services/stt/deepgram), [Fal Wizper](https://docs.pipecat.ai/server/services/stt/fal), [Gladia](https://docs.pipecat.ai/server/services/stt/gladia), [Google](https://docs.pipecat.ai/server/services/stt/google), [Groq (Whisper)](https://docs.pipecat.ai/server/services/stt/groq), [NVIDIA Riva](https://docs.pipecat.ai/server/services/stt/riva), [OpenAI (Whisper)](https://docs.pipecat.ai/server/services/stt/openai), [SambaNova (Whisper)](https://docs.pipecat.ai/server/services/stt/sambanova), [Soniox](https://docs.pipecat.ai/server/services/stt/soniox), [Speechmatics](https://docs.pipecat.ai/server/services/stt/speechmatics), [Ultravox](https://docs.pipecat.ai/server/services/stt/ultravox), [Whisper](https://docs.pipecat.ai/server/services/stt/whisper) |
|
||||
| LLMs | [Anthropic](https://docs.pipecat.ai/server/services/llm/anthropic), [AWS](https://docs.pipecat.ai/server/services/llm/aws), [Azure](https://docs.pipecat.ai/server/services/llm/azure), [Cerebras](https://docs.pipecat.ai/server/services/llm/cerebras), [DeepSeek](https://docs.pipecat.ai/server/services/llm/deepseek), [Fireworks AI](https://docs.pipecat.ai/server/services/llm/fireworks), [Gemini](https://docs.pipecat.ai/server/services/llm/gemini), [Grok](https://docs.pipecat.ai/server/services/llm/grok), [Groq](https://docs.pipecat.ai/server/services/llm/groq), [NVIDIA NIM](https://docs.pipecat.ai/server/services/llm/nim), [Ollama](https://docs.pipecat.ai/server/services/llm/ollama), [OpenAI](https://docs.pipecat.ai/server/services/llm/openai), [OpenRouter](https://docs.pipecat.ai/server/services/llm/openrouter), [Perplexity](https://docs.pipecat.ai/server/services/llm/perplexity), [Qwen](https://docs.pipecat.ai/server/services/llm/qwen), [SambaNova](https://docs.pipecat.ai/server/services/llm/sambanova) [Together AI](https://docs.pipecat.ai/server/services/llm/together) |
|
||||
| LLMs | [Anthropic](https://docs.pipecat.ai/server/services/llm/anthropic), [AWS](https://docs.pipecat.ai/server/services/llm/aws), [Azure](https://docs.pipecat.ai/server/services/llm/azure), [Cerebras](https://docs.pipecat.ai/server/services/llm/cerebras), [DeepSeek](https://docs.pipecat.ai/server/services/llm/deepseek), [Fireworks AI](https://docs.pipecat.ai/server/services/llm/fireworks), [Gemini](https://docs.pipecat.ai/server/services/llm/gemini), [Grok](https://docs.pipecat.ai/server/services/llm/grok), [Groq](https://docs.pipecat.ai/server/services/llm/groq), [Mistral](https://docs.pipecat.ai/server/services/llm/mistral), [NVIDIA NIM](https://docs.pipecat.ai/server/services/llm/nim), [Ollama](https://docs.pipecat.ai/server/services/llm/ollama), [OpenAI](https://docs.pipecat.ai/server/services/llm/openai), [OpenRouter](https://docs.pipecat.ai/server/services/llm/openrouter), [Perplexity](https://docs.pipecat.ai/server/services/llm/perplexity), [Qwen](https://docs.pipecat.ai/server/services/llm/qwen), [SambaNova](https://docs.pipecat.ai/server/services/llm/sambanova) [Together AI](https://docs.pipecat.ai/server/services/llm/together) |
|
||||
| Text-to-Speech | [Async](https://docs.pipecat.ai/server/services/tts/asyncai), [AWS](https://docs.pipecat.ai/server/services/tts/aws), [Azure](https://docs.pipecat.ai/server/services/tts/azure), [Cartesia](https://docs.pipecat.ai/server/services/tts/cartesia), [Deepgram](https://docs.pipecat.ai/server/services/tts/deepgram), [ElevenLabs](https://docs.pipecat.ai/server/services/tts/elevenlabs), [Fish](https://docs.pipecat.ai/server/services/tts/fish), [Google](https://docs.pipecat.ai/server/services/tts/google), [Groq](https://docs.pipecat.ai/server/services/tts/groq), [Inworld](https://docs.pipecat.ai/server/services/tts/inworld), [LMNT](https://docs.pipecat.ai/server/services/tts/lmnt), [MiniMax](https://docs.pipecat.ai/server/services/tts/minimax), [Neuphonic](https://docs.pipecat.ai/server/services/tts/neuphonic), [NVIDIA Riva](https://docs.pipecat.ai/server/services/tts/riva), [OpenAI](https://docs.pipecat.ai/server/services/tts/openai), [Piper](https://docs.pipecat.ai/server/services/tts/piper), [PlayHT](https://docs.pipecat.ai/server/services/tts/playht), [Rime](https://docs.pipecat.ai/server/services/tts/rime), [Sarvam](https://docs.pipecat.ai/server/services/tts/sarvam), [XTTS](https://docs.pipecat.ai/server/services/tts/xtts) |
|
||||
| Speech-to-Speech | [AWS Nova Sonic](https://docs.pipecat.ai/server/services/s2s/aws), [Gemini Multimodal Live](https://docs.pipecat.ai/server/services/s2s/gemini), [OpenAI Realtime](https://docs.pipecat.ai/server/services/s2s/openai) |
|
||||
| Transport | [Daily (WebRTC)](https://docs.pipecat.ai/server/services/transport/daily), [FastAPI Websocket](https://docs.pipecat.ai/server/services/transport/fastapi-websocket), [SmallWebRTCTransport](https://docs.pipecat.ai/server/services/transport/small-webrtc), [WebSocket Server](https://docs.pipecat.ai/server/services/transport/websocket-server), Local |
|
||||
|
||||
165
examples/foundational/14w-function-calling-mistral.py
Normal file
165
examples/foundational/14w-function-calling-mistral.py
Normal file
@@ -0,0 +1,165 @@
|
||||
#
|
||||
# Copyright (c) 2024–2025, Daily
|
||||
#
|
||||
# SPDX-License-Identifier: BSD 2-Clause License
|
||||
#
|
||||
|
||||
import os
|
||||
|
||||
from dotenv import load_dotenv
|
||||
from loguru import logger
|
||||
|
||||
from pipecat.adapters.schemas.function_schema import FunctionSchema
|
||||
from pipecat.adapters.schemas.tools_schema import ToolsSchema
|
||||
from pipecat.audio.vad.silero import SileroVADAnalyzer
|
||||
from pipecat.frames.frames import TTSSpeakFrame
|
||||
from pipecat.pipeline.pipeline import Pipeline
|
||||
from pipecat.pipeline.runner import PipelineRunner
|
||||
from pipecat.pipeline.task import PipelineParams, PipelineTask
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
from pipecat.runner.types import RunnerArguments
|
||||
from pipecat.runner.utils import create_transport
|
||||
from pipecat.services.cartesia.tts import CartesiaTTSService
|
||||
from pipecat.services.deepgram.stt import DeepgramSTTService
|
||||
from pipecat.services.llm_service import FunctionCallParams
|
||||
from pipecat.services.mistral.llm import MistralLLMService
|
||||
from pipecat.transports.base_transport import BaseTransport, TransportParams
|
||||
from pipecat.transports.network.fastapi_websocket import FastAPIWebsocketParams
|
||||
from pipecat.transports.services.daily import DailyParams
|
||||
|
||||
load_dotenv(override=True)
|
||||
|
||||
|
||||
async def fetch_weather_from_api(params: FunctionCallParams):
|
||||
await params.result_callback({"conditions": "nice", "temperature": "75"})
|
||||
|
||||
|
||||
async def fetch_restaurant_recommendation(params: FunctionCallParams):
|
||||
await params.result_callback({"name": "The Golden Dragon"})
|
||||
|
||||
|
||||
# We store functions so objects (e.g. SileroVADAnalyzer) don't get
|
||||
# instantiated. The function will be called when the desired transport gets
|
||||
# selected.
|
||||
transport_params = {
|
||||
"daily": lambda: DailyParams(
|
||||
audio_in_enabled=True,
|
||||
audio_out_enabled=True,
|
||||
vad_analyzer=SileroVADAnalyzer(),
|
||||
),
|
||||
"twilio": lambda: FastAPIWebsocketParams(
|
||||
audio_in_enabled=True,
|
||||
audio_out_enabled=True,
|
||||
vad_analyzer=SileroVADAnalyzer(),
|
||||
),
|
||||
"webrtc": lambda: TransportParams(
|
||||
audio_in_enabled=True,
|
||||
audio_out_enabled=True,
|
||||
vad_analyzer=SileroVADAnalyzer(),
|
||||
),
|
||||
}
|
||||
|
||||
|
||||
async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
|
||||
logger.info(f"Starting bot")
|
||||
|
||||
stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
|
||||
|
||||
tts = CartesiaTTSService(
|
||||
api_key=os.getenv("CARTESIA_API_KEY"),
|
||||
voice_id="71a7ad14-091c-4e8e-a314-022ece01c121", # British Reading Lady
|
||||
)
|
||||
|
||||
llm = MistralLLMService(api_key=os.getenv("MISTRAL_API_KEY"))
|
||||
|
||||
# You can also register a function_name of None to get all functions
|
||||
# sent to the same callback with an additional function_name parameter.
|
||||
llm.register_function("get_current_weather", fetch_weather_from_api)
|
||||
llm.register_function("get_restaurant_recommendation", fetch_restaurant_recommendation)
|
||||
|
||||
weather_function = FunctionSchema(
|
||||
name="get_current_weather",
|
||||
description="Get the current weather",
|
||||
properties={
|
||||
"location": {
|
||||
"type": "string",
|
||||
"description": "The city and state, e.g. San Francisco, CA",
|
||||
},
|
||||
"format": {
|
||||
"type": "string",
|
||||
"enum": ["celsius", "fahrenheit"],
|
||||
"description": "The temperature unit to use. Infer this from the user's location.",
|
||||
},
|
||||
},
|
||||
required=["location", "format"],
|
||||
)
|
||||
restaurant_function = FunctionSchema(
|
||||
name="get_restaurant_recommendation",
|
||||
description="Get a restaurant recommendation",
|
||||
properties={
|
||||
"location": {
|
||||
"type": "string",
|
||||
"description": "The city and state, e.g. San Francisco, CA",
|
||||
},
|
||||
},
|
||||
required=["location"],
|
||||
)
|
||||
tools = ToolsSchema(standard_tools=[weather_function, restaurant_function])
|
||||
|
||||
messages = [
|
||||
{
|
||||
"role": "system",
|
||||
"content": "You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. Your output will be converted to audio so don't include special characters in your answers. Respond to what the user said in a creative and helpful way.",
|
||||
},
|
||||
]
|
||||
|
||||
context = OpenAILLMContext(messages, tools)
|
||||
context_aggregator = llm.create_context_aggregator(context)
|
||||
|
||||
pipeline = Pipeline(
|
||||
[
|
||||
transport.input(),
|
||||
stt,
|
||||
context_aggregator.user(),
|
||||
llm,
|
||||
tts,
|
||||
transport.output(),
|
||||
context_aggregator.assistant(),
|
||||
]
|
||||
)
|
||||
|
||||
task = PipelineTask(
|
||||
pipeline,
|
||||
params=PipelineParams(
|
||||
enable_metrics=True,
|
||||
enable_usage_metrics=True,
|
||||
),
|
||||
idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
|
||||
)
|
||||
|
||||
@transport.event_handler("on_client_connected")
|
||||
async def on_client_connected(transport, client):
|
||||
logger.info(f"Client connected")
|
||||
# Kick off the conversation.
|
||||
await task.queue_frames([context_aggregator.user().get_context_frame()])
|
||||
|
||||
@transport.event_handler("on_client_disconnected")
|
||||
async def on_client_disconnected(transport, client):
|
||||
logger.info(f"Client disconnected")
|
||||
await task.cancel()
|
||||
|
||||
runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
|
||||
|
||||
await runner.run(task)
|
||||
|
||||
|
||||
async def bot(runner_args: RunnerArguments):
|
||||
"""Main bot entry point compatible with Pipecat Cloud."""
|
||||
transport = await create_transport(runner_args, transport_params)
|
||||
await run_bot(transport, runner_args)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
from pipecat.runner.run import main
|
||||
|
||||
main()
|
||||
@@ -73,6 +73,7 @@ lmnt = [ "websockets>=13.1,<15.0" ]
|
||||
local = [ "pyaudio~=0.2.14" ]
|
||||
mcp = [ "mcp[cli]~=1.9.4" ]
|
||||
mem0 = [ "mem0ai~=0.1.94" ]
|
||||
mistral = []
|
||||
mlx-whisper = [ "mlx-whisper~=0.4.2" ]
|
||||
moondream = [ "accelerate~=1.10.0", "einops~=0.8.0", "pyvips[binary]~=3.0.0", "timm~=1.0.13", "transformers>=4.48.0" ]
|
||||
nim = []
|
||||
|
||||
@@ -113,6 +113,7 @@ TESTS_14 = [
|
||||
("14q-function-calling-qwen.py", PROMPT_WEATHER, EVAL_WEATHER),
|
||||
("14r-function-calling-aws.py", PROMPT_WEATHER, EVAL_WEATHER),
|
||||
("14v-function-calling-openai.py", PROMPT_WEATHER, EVAL_WEATHER),
|
||||
("14w-function-calling-mistral.py", PROMPT_WEATHER, EVAL_WEATHER),
|
||||
# Currently not working.
|
||||
# ("14c-function-calling-together.py", PROMPT_WEATHER, EVAL_WEATHER),
|
||||
# ("14k-function-calling-cerebras.py", PROMPT_WEATHER, EVAL_WEATHER),
|
||||
|
||||
0
src/pipecat/services/mistral/__init__.py
Normal file
0
src/pipecat/services/mistral/__init__.py
Normal file
182
src/pipecat/services/mistral/llm.py
Normal file
182
src/pipecat/services/mistral/llm.py
Normal file
@@ -0,0 +1,182 @@
|
||||
#
|
||||
# Copyright (c) 2024–2025, Daily
|
||||
#
|
||||
# SPDX-License-Identifier: BSD 2-Clause License
|
||||
#
|
||||
|
||||
"""Mistral LLM service implementation using OpenAI-compatible interface."""
|
||||
|
||||
from typing import List, Sequence
|
||||
|
||||
from loguru import logger
|
||||
from openai import AsyncStream
|
||||
from openai.types.chat import ChatCompletionChunk, ChatCompletionMessageParam
|
||||
|
||||
from pipecat.frames.frames import FunctionCallFromLLM
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
from pipecat.services.openai.llm import OpenAILLMService
|
||||
|
||||
|
||||
class MistralLLMService(OpenAILLMService):
|
||||
"""A service for interacting with Mistral's API using the OpenAI-compatible interface.
|
||||
|
||||
This service extends OpenAILLMService to connect to Mistral's API endpoint while
|
||||
maintaining full compatibility with OpenAI's interface and functionality.
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
*,
|
||||
api_key: str,
|
||||
base_url: str = "https://api.mistral.ai/v1",
|
||||
model: str = "mistral-small-latest",
|
||||
**kwargs,
|
||||
):
|
||||
"""Initialize the Mistral LLM service.
|
||||
|
||||
Args:
|
||||
api_key: The API key for accessing Mistral's API.
|
||||
base_url: The base URL for Mistral API. Defaults to "https://api.mistral.ai/v1".
|
||||
model: The model identifier to use. Defaults to "mistral-small-latest".
|
||||
**kwargs: Additional keyword arguments passed to OpenAILLMService.
|
||||
"""
|
||||
super().__init__(api_key=api_key, base_url=base_url, model=model, **kwargs)
|
||||
|
||||
def create_client(self, api_key=None, base_url=None, **kwargs):
|
||||
"""Create OpenAI-compatible client for Mistral API endpoint.
|
||||
|
||||
Args:
|
||||
api_key: The API key for authentication. If None, uses instance key.
|
||||
base_url: The base URL for the API. If None, uses instance URL.
|
||||
**kwargs: Additional arguments passed to the client constructor.
|
||||
|
||||
Returns:
|
||||
An OpenAI-compatible client configured for Mistral API.
|
||||
"""
|
||||
logger.debug(f"Creating Mistral client with api {base_url}")
|
||||
return super().create_client(api_key, base_url, **kwargs)
|
||||
|
||||
def _apply_mistral_assistant_prefix(
|
||||
self, messages: List[ChatCompletionMessageParam]
|
||||
) -> List[ChatCompletionMessageParam]:
|
||||
"""Apply Mistral's assistant message prefix requirement.
|
||||
|
||||
Mistral requires assistant messages to have prefix=True when they
|
||||
are the final message in a conversation. According to Mistral's API:
|
||||
- Assistant messages with prefix=True MUST be the last message
|
||||
- Only add prefix=True to the final assistant message when needed
|
||||
- This allows assistant messages to be accepted as the last message
|
||||
|
||||
Args:
|
||||
messages: The original list of messages.
|
||||
|
||||
Returns:
|
||||
Messages with Mistral prefix requirement applied to final assistant message.
|
||||
"""
|
||||
if not messages:
|
||||
return messages
|
||||
|
||||
# Create a copy to avoid modifying the original
|
||||
fixed_messages = [dict(msg) for msg in messages]
|
||||
|
||||
# Get the last message
|
||||
last_message = fixed_messages[-1]
|
||||
|
||||
# Only add prefix=True to the last message if it's an assistant message
|
||||
# and Mistral would otherwise reject it
|
||||
if last_message.get("role") == "assistant" and "prefix" not in last_message:
|
||||
last_message["prefix"] = True
|
||||
|
||||
return fixed_messages
|
||||
|
||||
async def run_function_calls(self, function_calls: Sequence[FunctionCallFromLLM]):
|
||||
"""Execute function calls, filtering out already-completed ones.
|
||||
|
||||
Mistral and OpenAI have different function call detection patterns:
|
||||
|
||||
OpenAI (Stream-based detection):
|
||||
- Detects function calls only from streaming chunks as the LLM generates them
|
||||
- Second LLM completion doesn't re-detect existing tool_calls in message history
|
||||
- Function calls execute exactly once
|
||||
|
||||
Mistral (Message-based detection):
|
||||
- Detects function calls from the complete message history on each completion
|
||||
- Second LLM completion with the response re-detects the same tool_calls from
|
||||
previous messages
|
||||
- Without filtering, function calls would execute twice
|
||||
|
||||
This method prevents duplicate execution by:
|
||||
1. Checking message history for existing tool result messages
|
||||
2. Filtering out function calls that already have corresponding results
|
||||
3. Only executing function calls that haven't been completed yet
|
||||
|
||||
Note: This filtering prevents duplicate function execution, but the
|
||||
on_function_calls_started event may still fire twice due to the detection
|
||||
pattern difference. This is expected behavior.
|
||||
|
||||
Args:
|
||||
function_calls: The function calls to potentially execute.
|
||||
"""
|
||||
if not function_calls:
|
||||
return
|
||||
|
||||
# Filter out function calls that already have results
|
||||
calls_to_execute = []
|
||||
|
||||
# Get messages from the first function call's context (they should all have the same context)
|
||||
messages = function_calls[0].context.get_messages() if function_calls else []
|
||||
|
||||
# Get all tool_call_ids that already have results
|
||||
executed_call_ids = set()
|
||||
for msg in messages:
|
||||
if msg.get("role") == "tool" and msg.get("tool_call_id"):
|
||||
executed_call_ids.add(msg.get("tool_call_id"))
|
||||
|
||||
# Only include function calls that haven't been executed yet
|
||||
for call in function_calls:
|
||||
if call.tool_call_id not in executed_call_ids:
|
||||
calls_to_execute.append(call)
|
||||
else:
|
||||
logger.trace(
|
||||
f"Skipping already-executed function call: {call.function_name}:{call.tool_call_id}"
|
||||
)
|
||||
|
||||
# Call parent method with filtered list
|
||||
if calls_to_execute:
|
||||
await super().run_function_calls(calls_to_execute)
|
||||
|
||||
def build_chat_completion_params(
|
||||
self, context: OpenAILLMContext, messages: List[ChatCompletionMessageParam]
|
||||
) -> dict:
|
||||
"""Build parameters for Mistral chat completion request.
|
||||
|
||||
Handles Mistral-specific requirements including:
|
||||
- Assistant message prefix requirement for API compatibility
|
||||
- Parameter mapping (random_seed instead of seed)
|
||||
- Core completion settings
|
||||
"""
|
||||
# Apply Mistral's assistant prefix requirement for API compatibility
|
||||
fixed_messages = self._apply_mistral_assistant_prefix(messages)
|
||||
|
||||
params = {
|
||||
"model": self.model_name,
|
||||
"stream": True,
|
||||
"messages": fixed_messages,
|
||||
"tools": context.tools,
|
||||
"tool_choice": context.tool_choice,
|
||||
"frequency_penalty": self._settings["frequency_penalty"],
|
||||
"presence_penalty": self._settings["presence_penalty"],
|
||||
"temperature": self._settings["temperature"],
|
||||
"top_p": self._settings["top_p"],
|
||||
"max_tokens": self._settings["max_tokens"],
|
||||
}
|
||||
|
||||
# Handle Mistral-specific parameter mapping
|
||||
# Mistral uses "random_seed" instead of "seed"
|
||||
if self._settings["seed"]:
|
||||
params["random_seed"] = self._settings["seed"]
|
||||
|
||||
# Add any extra parameters
|
||||
params.update(self._settings["extra"])
|
||||
|
||||
return params
|
||||
Reference in New Issue
Block a user