pydantic_ai.models.groq

Setup

For details on how to set up authentication with this model, see model configuration for Groq.

GroqModelName module-attribute

GroqModelName = Literal[
    "llama-3.3-70b-versatile",
    "llama-3.1-70b-versatile",
    "llama3-groq-70b-8192-tool-use-preview",
    "llama3-groq-8b-8192-tool-use-preview",
    "llama-3.1-70b-specdec",
    "llama-3.1-8b-instant",
    "llama-3.2-1b-preview",
    "llama-3.2-3b-preview",
    "llama-3.2-11b-vision-preview",
    "llama-3.2-90b-vision-preview",
    "llama3-70b-8192",
    "llama3-8b-8192",
    "mixtral-8x7b-32768",
    "gemma2-9b-it",
    "gemma-7b-it",
]

Named Groq models.

See the Groq docs for a full list.

GroqModel dataclass

Bases: Model

A model that uses the Groq API.

Internally, this uses the Groq Python client to interact with the API.

Apart from __init__, all methods are private or match those of the base class.

Source code in pydantic_ai_slim/pydantic_ai/models/groq.py

@dataclass(init=False)
class GroqModel(Model):
    """A model that uses the Groq API.

    Internally, this uses the [Groq Python client](https://github.com/groq/groq-python) to interact with the API.

    Apart from `__init__`, all methods are private or match those of the base class.
    """

    model_name: GroqModelName
    client: AsyncGroq = field(repr=False)

    def __init__(
        self,
        model_name: GroqModelName,
        *,
        api_key: str | None = None,
        groq_client: AsyncGroq | None = None,
        http_client: AsyncHTTPClient | None = None,
    ):
        """Initialize a Groq model.

        Args:
            model_name: The name of the Groq model to use. List of model names available
                [here](https://console.groq.com/docs/models).
            api_key: The API key to use for authentication, if not provided, the `GROQ_API_KEY` environment variable
                will be used if available.
            groq_client: An existing
                [`AsyncGroq`](https://github.com/groq/groq-python?tab=readme-ov-file#async-usage)
                client to use, if provided, `api_key` and `http_client` must be `None`.
            http_client: An existing `httpx.AsyncClient` to use for making HTTP requests.
        """
        self.model_name = model_name
        if groq_client is not None:
            assert http_client is None, 'Cannot provide both `groq_client` and `http_client`'
            assert api_key is None, 'Cannot provide both `groq_client` and `api_key`'
            self.client = groq_client
        elif http_client is not None:
            self.client = AsyncGroq(api_key=api_key, http_client=http_client)
        else:
            self.client = AsyncGroq(api_key=api_key, http_client=cached_async_http_client())

    async def agent_model(
        self,
        *,
        function_tools: list[ToolDefinition],
        allow_text_result: bool,
        result_tools: list[ToolDefinition],
    ) -> AgentModel:
        check_allow_model_requests()
        tools = [self._map_tool_definition(r) for r in function_tools]
        if result_tools:
            tools += [self._map_tool_definition(r) for r in result_tools]
        return GroqAgentModel(
            self.client,
            self.model_name,
            allow_text_result,
            tools,
        )

    def name(self) -> str:
        return f'groq:{self.model_name}'

    @staticmethod
    def _map_tool_definition(f: ToolDefinition) -> chat.ChatCompletionToolParam:
        return {
            'type': 'function',
            'function': {
                'name': f.name,
                'description': f.description,
                'parameters': f.parameters_json_schema,
            },
        }

__init__

__init__(
    model_name: GroqModelName,
    *,
    api_key: str | None = None,
    groq_client: AsyncGroq | None = None,
    http_client: AsyncClient | None = None
)

Initialize a Groq model.

Parameters:

Name	Type	Description	Default
model_name	GroqModelName	The name of the Groq model to use. List of model names available here.	required
api_key	str | None	The API key to use for authentication, if not provided, the GROQ_API_KEY environment variable will be used if available.	None
groq_client	AsyncGroq | None	An existing AsyncGroq client to use, if provided, api_key and http_client must be None.	None
http_client	AsyncClient | None	An existing httpx.AsyncClient to use for making HTTP requests.	None

Source code in pydantic_ai_slim/pydantic_ai/models/groq.py

def __init__(
    self,
    model_name: GroqModelName,
    *,
    api_key: str | None = None,
    groq_client: AsyncGroq | None = None,
    http_client: AsyncHTTPClient | None = None,
):
    """Initialize a Groq model.

    Args:
        model_name: The name of the Groq model to use. List of model names available
            [here](https://console.groq.com/docs/models).
        api_key: The API key to use for authentication, if not provided, the `GROQ_API_KEY` environment variable
            will be used if available.
        groq_client: An existing
            [`AsyncGroq`](https://github.com/groq/groq-python?tab=readme-ov-file#async-usage)
            client to use, if provided, `api_key` and `http_client` must be `None`.
        http_client: An existing `httpx.AsyncClient` to use for making HTTP requests.
    """
    self.model_name = model_name
    if groq_client is not None:
        assert http_client is None, 'Cannot provide both `groq_client` and `http_client`'
        assert api_key is None, 'Cannot provide both `groq_client` and `api_key`'
        self.client = groq_client
    elif http_client is not None:
        self.client = AsyncGroq(api_key=api_key, http_client=http_client)
    else:
        self.client = AsyncGroq(api_key=api_key, http_client=cached_async_http_client())

GroqAgentModel dataclass

Bases: AgentModel

Implementation of AgentModel for Groq models.

Source code in pydantic_ai_slim/pydantic_ai/models/groq.py

@dataclass
class GroqAgentModel(AgentModel):
    """Implementation of `AgentModel` for Groq models."""

    client: AsyncGroq
    model_name: str
    allow_text_result: bool
    tools: list[chat.ChatCompletionToolParam]

    async def request(
        self, messages: list[ModelMessage], model_settings: ModelSettings | None
    ) -> tuple[ModelResponse, usage.Usage]:
        response = await self._completions_create(messages, False, model_settings)
        return self._process_response(response), _map_usage(response)

    @asynccontextmanager
    async def request_stream(
        self, messages: list[ModelMessage], model_settings: ModelSettings | None
    ) -> AsyncIterator[StreamedResponse]:
        response = await self._completions_create(messages, True, model_settings)
        async with response:
            yield await self._process_streamed_response(response)

    @overload
    async def _completions_create(
        self, messages: list[ModelMessage], stream: Literal[True], model_settings: ModelSettings | None
    ) -> AsyncStream[ChatCompletionChunk]:
        pass

    @overload
    async def _completions_create(
        self, messages: list[ModelMessage], stream: Literal[False], model_settings: ModelSettings | None
    ) -> chat.ChatCompletion:
        pass

    async def _completions_create(
        self, messages: list[ModelMessage], stream: bool, model_settings: ModelSettings | None
    ) -> chat.ChatCompletion | AsyncStream[ChatCompletionChunk]:
        # standalone function to make it easier to override
        if not self.tools:
            tool_choice: Literal['none', 'required', 'auto'] | None = None
        elif not self.allow_text_result:
            tool_choice = 'required'
        else:
            tool_choice = 'auto'

        groq_messages = list(chain(*(self._map_message(m) for m in messages)))

        model_settings = model_settings or {}

        return await self.client.chat.completions.create(
            model=str(self.model_name),
            messages=groq_messages,
            n=1,
            parallel_tool_calls=True if self.tools else NOT_GIVEN,
            tools=self.tools or NOT_GIVEN,
            tool_choice=tool_choice or NOT_GIVEN,
            stream=stream,
            max_tokens=model_settings.get('max_tokens', NOT_GIVEN),
            temperature=model_settings.get('temperature', NOT_GIVEN),
            top_p=model_settings.get('top_p', NOT_GIVEN),
            timeout=model_settings.get('timeout', NOT_GIVEN),
        )

    @staticmethod
    def _process_response(response: chat.ChatCompletion) -> ModelResponse:
        """Process a non-streamed response, and prepare a message to return."""
        timestamp = datetime.fromtimestamp(response.created, tz=timezone.utc)
        choice = response.choices[0]
        items: list[ModelResponsePart] = []
        if choice.message.content is not None:
            items.append(TextPart(content=choice.message.content))
        if choice.message.tool_calls is not None:
            for c in choice.message.tool_calls:
                items.append(
                    ToolCallPart.from_raw_args(tool_name=c.function.name, args=c.function.arguments, tool_call_id=c.id)
                )
        return ModelResponse(items, timestamp=timestamp)

    @staticmethod
    async def _process_streamed_response(response: AsyncStream[ChatCompletionChunk]) -> GroqStreamedResponse:
        """Process a streamed response, and prepare a streaming response to return."""
        peekable_response = _utils.PeekableAsyncStream(response)
        first_chunk = await peekable_response.peek()
        if isinstance(first_chunk, _utils.Unset):
            raise UnexpectedModelBehavior('Streamed response ended without content or tool calls')

        return GroqStreamedResponse(peekable_response, datetime.fromtimestamp(first_chunk.created, tz=timezone.utc))

    @classmethod
    def _map_message(cls, message: ModelMessage) -> Iterable[chat.ChatCompletionMessageParam]:
        """Just maps a `pydantic_ai.Message` to a `groq.types.ChatCompletionMessageParam`."""
        if isinstance(message, ModelRequest):
            yield from cls._map_user_message(message)
        elif isinstance(message, ModelResponse):
            texts: list[str] = []
            tool_calls: list[chat.ChatCompletionMessageToolCallParam] = []
            for item in message.parts:
                if isinstance(item, TextPart):
                    texts.append(item.content)
                elif isinstance(item, ToolCallPart):
                    tool_calls.append(_map_tool_call(item))
                else:
                    assert_never(item)
            message_param = chat.ChatCompletionAssistantMessageParam(role='assistant')
            if texts:
                # Note: model responses from this model should only have one text item, so the following
                # shouldn't merge multiple texts into one unless you switch models between runs:
                message_param['content'] = '\n\n'.join(texts)
            if tool_calls:
                message_param['tool_calls'] = tool_calls
            yield message_param
        else:
            assert_never(message)

    @classmethod
    def _map_user_message(cls, message: ModelRequest) -> Iterable[chat.ChatCompletionMessageParam]:
        for part in message.parts:
            if isinstance(part, SystemPromptPart):
                yield chat.ChatCompletionSystemMessageParam(role='system', content=part.content)
            elif isinstance(part, UserPromptPart):
                yield chat.ChatCompletionUserMessageParam(role='user', content=part.content)
            elif isinstance(part, ToolReturnPart):
                yield chat.ChatCompletionToolMessageParam(
                    role='tool',
                    tool_call_id=_guard_tool_call_id(t=part, model_source='Groq'),
                    content=part.model_response_str(),
                )
            elif isinstance(part, RetryPromptPart):
                if part.tool_name is None:
                    yield chat.ChatCompletionUserMessageParam(role='user', content=part.model_response())
                else:
                    yield chat.ChatCompletionToolMessageParam(
                        role='tool',
                        tool_call_id=_guard_tool_call_id(t=part, model_source='Groq'),
                        content=part.model_response(),
                    )

GroqStreamedResponse dataclass

Bases: StreamedResponse

Implementation of StreamedResponse for Groq models.

Source code in pydantic_ai_slim/pydantic_ai/models/groq.py

@dataclass
class GroqStreamedResponse(StreamedResponse):
    """Implementation of `StreamedResponse` for Groq models."""

    _response: AsyncIterable[ChatCompletionChunk]
    _timestamp: datetime

    async def _get_event_iterator(self) -> AsyncIterator[ModelResponseStreamEvent]:
        async for chunk in self._response:
            self._usage += _map_usage(chunk)

            try:
                choice = chunk.choices[0]
            except IndexError:
                continue

            # Handle the text part of the response
            content = choice.delta.content
            if content is not None:
                yield self._parts_manager.handle_text_delta(vendor_part_id='content', content=content)

            # Handle the tool calls
            for dtc in choice.delta.tool_calls or []:
                maybe_event = self._parts_manager.handle_tool_call_delta(
                    vendor_part_id=dtc.index,
                    tool_name=dtc.function and dtc.function.name,
                    args=dtc.function and dtc.function.arguments,
                    tool_call_id=dtc.id,
                )
                if maybe_event is not None:
                    yield maybe_event

    def timestamp(self) -> datetime:
        return self._timestamp