pydantic-ai-slim 0.0.11__py3-none-any.whl → 0.0.13__py3-none-any.whl

pydantic_ai/models/anthropic.py

@@ -0,0 +1,344 @@
+from __future__ import annotations as _annotations
+
+from collections.abc import AsyncIterator
+from contextlib import asynccontextmanager
+from dataclasses import dataclass, field
+from typing import Any, Literal, Union, cast, overload
+
+from httpx import AsyncClient as AsyncHTTPClient
+from typing_extensions import assert_never
+
+from .. import result
+from .._utils import guard_tool_call_id as _guard_tool_call_id
+from ..messages import (
+    ArgsDict,
+    ModelMessage,
+    ModelRequest,
+    ModelResponse,
+    ModelResponsePart,
+    RetryPromptPart,
+    SystemPromptPart,
+    TextPart,
+    ToolCallPart,
+    ToolReturnPart,
+    UserPromptPart,
+)
+from ..settings import ModelSettings
+from ..tools import ToolDefinition
+from . import (
+    AgentModel,
+    EitherStreamedResponse,
+    Model,
+    cached_async_http_client,
+    check_allow_model_requests,
+)
+
+try:
+    from anthropic import NOT_GIVEN, AsyncAnthropic, AsyncStream
+    from anthropic.types import (
+        Message as AnthropicMessage,
+        MessageParam,
+        RawMessageDeltaEvent,
+        RawMessageStartEvent,
+        RawMessageStreamEvent,
+        TextBlock,
+        TextBlockParam,
+        ToolChoiceParam,
+        ToolParam,
+        ToolResultBlockParam,
+        ToolUseBlock,
+        ToolUseBlockParam,
+    )
+except ImportError as _import_error:
+    raise ImportError(
+        'Please install `anthropic` to use the Anthropic model, '
+        "you can use the `anthropic` optional group — `pip install 'pydantic-ai-slim[anthropic]'`"
+    ) from _import_error
+
+LatestAnthropicModelNames = Literal[
+    'claude-3-5-haiku-latest',
+    'claude-3-5-sonnet-latest',
+    'claude-3-opus-latest',
+]
+"""Latest named Anthropic models."""
+
+AnthropicModelName = Union[str, LatestAnthropicModelNames]
+"""Possible Anthropic model names.
+
+Since Anthropic supports a variety of date-stamped models, we explicitly list the latest models but
+allow any name in the type hints.
+Since [the Anthropic docs](https://docs.anthropic.com/en/docs/about-claude/models) for a full list.
+"""
+
+
+@dataclass(init=False)
+class AnthropicModel(Model):
+    """A model that uses the Anthropic API.
+
+    Internally, this uses the [Anthropic Python client](https://github.com/anthropics/anthropic-sdk-python) to interact with the API.
+
+    Apart from `__init__`, all methods are private or match those of the base class.
+
+    !!! note
+        The `AnthropicModel` class does not yet support streaming responses.
+        We anticipate adding support for streaming responses in a near-term future release.
+    """
+
+    model_name: AnthropicModelName
+    client: AsyncAnthropic = field(repr=False)
+
+    def __init__(
+        self,
+        model_name: AnthropicModelName,
+        *,
+        api_key: str | None = None,
+        anthropic_client: AsyncAnthropic | None = None,
+        http_client: AsyncHTTPClient | None = None,
+    ):
+        """Initialize an Anthropic model.
+
+        Args:
+            model_name: The name of the Anthropic model to use. List of model names available
+                [here](https://docs.anthropic.com/en/docs/about-claude/models).
+            api_key: The API key to use for authentication, if not provided, the `ANTHROPIC_API_KEY` environment variable
+                will be used if available.
+            anthropic_client: An existing
+                [`AsyncAnthropic`](https://github.com/anthropics/anthropic-sdk-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 anthropic_client is not None:
+            assert http_client is None, 'Cannot provide both `anthropic_client` and `http_client`'
+            assert api_key is None, 'Cannot provide both `anthropic_client` and `api_key`'
+            self.client = anthropic_client
+        elif http_client is not None:
+            self.client = AsyncAnthropic(api_key=api_key, http_client=http_client)
+        else:
+            self.client = AsyncAnthropic(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 AnthropicAgentModel(
+            self.client,
+            self.model_name,
+            allow_text_result,
+            tools,
+        )
+
+    def name(self) -> str:
+        return self.model_name
+
+    @staticmethod
+    def _map_tool_definition(f: ToolDefinition) -> ToolParam:
+        return {
+            'name': f.name,
+            'description': f.description,
+            'input_schema': f.parameters_json_schema,
+        }
+
+
+@dataclass
+class AnthropicAgentModel(AgentModel):
+    """Implementation of `AgentModel` for Anthropic models."""
+
+    client: AsyncAnthropic
+    model_name: str
+    allow_text_result: bool
+    tools: list[ToolParam]
+
+    async def request(
+        self, messages: list[ModelMessage], model_settings: ModelSettings | None
+    ) -> tuple[ModelResponse, result.Cost]:
+        response = await self._messages_create(messages, False, model_settings)
+        return self._process_response(response), _map_cost(response)
+
+    @asynccontextmanager
+    async def request_stream(
+        self, messages: list[ModelMessage], model_settings: ModelSettings | None
+    ) -> AsyncIterator[EitherStreamedResponse]:
+        response = await self._messages_create(messages, True, model_settings)
+        async with response:
+            yield await self._process_streamed_response(response)
+
+    @overload
+    async def _messages_create(
+        self, messages: list[ModelMessage], stream: Literal[True], model_settings: ModelSettings | None
+    ) -> AsyncStream[RawMessageStreamEvent]:
+        pass
+
+    @overload
+    async def _messages_create(
+        self, messages: list[ModelMessage], stream: Literal[False], model_settings: ModelSettings | None
+    ) -> AnthropicMessage:
+        pass
+
+    async def _messages_create(
+        self, messages: list[ModelMessage], stream: bool, model_settings: ModelSettings | None
+    ) -> AnthropicMessage | AsyncStream[RawMessageStreamEvent]:
+        # standalone function to make it easier to override
+        if not self.tools:
+            tool_choice: ToolChoiceParam | None = None
+        elif not self.allow_text_result:
+            tool_choice = {'type': 'any'}
+        else:
+            tool_choice = {'type': 'auto'}
+
+        system_prompt, anthropic_messages = self._map_message(messages)
+
+        model_settings = model_settings or {}
+
+        return await self.client.messages.create(
+            max_tokens=model_settings.get('max_tokens', 1024),
+            system=system_prompt or NOT_GIVEN,
+            messages=anthropic_messages,
+            model=self.model_name,
+            tools=self.tools or NOT_GIVEN,
+            tool_choice=tool_choice or NOT_GIVEN,
+            stream=stream,
+            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: AnthropicMessage) -> ModelResponse:
+        """Process a non-streamed response, and prepare a message to return."""
+        items: list[ModelResponsePart] = []
+        for item in response.content:
+            if isinstance(item, TextBlock):
+                items.append(TextPart(item.text))
+            else:
+                assert isinstance(item, ToolUseBlock), 'unexpected item type'
+                items.append(
+                    ToolCallPart.from_dict(
+                        item.name,
+                        cast(dict[str, Any], item.input),
+                        item.id,
+                    )
+                )
+
+        return ModelResponse(items)
+
+    @staticmethod
+    async def _process_streamed_response(response: AsyncStream[RawMessageStreamEvent]) -> EitherStreamedResponse:
+        """TODO: Process a streamed response, and prepare a streaming response to return."""
+        # We don't yet support streamed responses from Anthropic, so we raise an error here for now.
+        # Streamed responses will be supported in a future release.
+
+        raise RuntimeError('Streamed responses are not yet supported for Anthropic models.')
+
+        # Should be returning some sort of AnthropicStreamTextResponse or AnthropicStreamStructuredResponse
+        # depending on the type of chunk we get, but we need to establish how we handle (and when we get) the following:
+        # RawMessageStartEvent
+        # RawMessageDeltaEvent
+        # RawMessageStopEvent
+        # RawContentBlockStartEvent
+        # RawContentBlockDeltaEvent
+        # RawContentBlockDeltaEvent
+        #
+        # We might refactor streaming internally before we implement this...
+
+    @staticmethod
+    def _map_message(messages: list[ModelMessage]) -> tuple[str, list[MessageParam]]:
+        """Just maps a `pydantic_ai.Message` to a `anthropic.types.MessageParam`."""
+        system_prompt: str = ''
+        anthropic_messages: list[MessageParam] = []
+        for m in messages:
+            if isinstance(m, ModelRequest):
+                for part in m.parts:
+                    if isinstance(part, SystemPromptPart):
+                        system_prompt += part.content
+                    elif isinstance(part, UserPromptPart):
+                        anthropic_messages.append(MessageParam(role='user', content=part.content))
+                    elif isinstance(part, ToolReturnPart):
+                        anthropic_messages.append(
+                            MessageParam(
+                                role='user',
+                                content=[
+                                    ToolResultBlockParam(
+                                        tool_use_id=_guard_tool_call_id(t=part, model_source='Anthropic'),
+                                        type='tool_result',
+                                        content=part.model_response_str(),
+                                        is_error=False,
+                                    )
+                                ],
+                            )
+                        )
+                    elif isinstance(part, RetryPromptPart):
+                        if part.tool_name is None:
+                            anthropic_messages.append(MessageParam(role='user', content=part.model_response()))
+                        else:
+                            anthropic_messages.append(
+                                MessageParam(
+                                    role='user',
+                                    content=[
+                                        ToolUseBlockParam(
+                                            id=_guard_tool_call_id(t=part, model_source='Anthropic'),
+                                            input=part.model_response(),
+                                            name=part.tool_name,
+                                            type='tool_use',
+                                        ),
+                                    ],
+                                )
+                            )
+            elif isinstance(m, ModelResponse):
+                content: list[TextBlockParam | ToolUseBlockParam] = []
+                for item in m.parts:
+                    if isinstance(item, TextPart):
+                        content.append(TextBlockParam(text=item.content, type='text'))
+                    else:
+                        assert isinstance(item, ToolCallPart)
+                        content.append(_map_tool_call(item))
+                anthropic_messages.append(MessageParam(role='assistant', content=content))
+            else:
+                assert_never(m)
+        return system_prompt, anthropic_messages
+
+
+def _map_tool_call(t: ToolCallPart) -> ToolUseBlockParam:
+    assert isinstance(t.args, ArgsDict), f'Expected ArgsDict, got {t.args}'
+    return ToolUseBlockParam(
+        id=_guard_tool_call_id(t=t, model_source='Anthropic'),
+        type='tool_use',
+        name=t.tool_name,
+        input=t.args.args_dict,
+    )
+
+
+def _map_cost(message: AnthropicMessage | RawMessageStreamEvent) -> result.Cost:
+    if isinstance(message, AnthropicMessage):
+        usage = message.usage
+    else:
+        if isinstance(message, RawMessageStartEvent):
+            usage = message.message.usage
+        elif isinstance(message, RawMessageDeltaEvent):
+            usage = message.usage
+        else:
+            # No usage information provided in:
+            # - RawMessageStopEvent
+            # - RawContentBlockStartEvent
+            # - RawContentBlockDeltaEvent
+            # - RawContentBlockStopEvent
+            usage = None
+
+    if usage is None:
+        return result.Cost()
+
+    request_tokens = getattr(usage, 'input_tokens', None)
+
+    return result.Cost(
+        # Usage coming from the RawMessageDeltaEvent doesn't have input token data, hence this getattr
+        request_tokens=request_tokens,
+        response_tokens=usage.output_tokens,
+        total_tokens=(request_tokens or 0) + usage.output_tokens,
+    )

pydantic_ai/models/function.py

@@ -2,9 +2,9 @@ from __future__ import annotations as _annotations
 
 import inspect
 import re
-from collections.abc import AsyncIterator, Awaitable, Iterable, Mapping, Sequence
+from collections.abc import AsyncIterator, Awaitable, Iterable
 from contextlib import asynccontextmanager
-from dataclasses import dataclass, field
+from dataclasses import dataclass, field, replace
 from datetime import datetime
 from itertools import chain
 from typing import Callable, Union, cast
@@ -13,15 +13,22 @@ import pydantic_core
 from typing_extensions import TypeAlias, assert_never, overload
 
 from .. import _utils, result
-from ..messages import ArgsJson, Message, ModelAnyResponse, ModelStructuredResponse, ToolCall
-from . import (
-    AbstractToolDefinition,
-    AgentModel,
-    EitherStreamedResponse,
-    Model,
-    StreamStructuredResponse,
-    StreamTextResponse,
+from ..messages import (
+    ArgsJson,
+    ModelMessage,
+    ModelRequest,
+    ModelResponse,
+    ModelResponsePart,
+    RetryPromptPart,
+    SystemPromptPart,
+    TextPart,
+    ToolCallPart,
+    ToolReturnPart,
+    UserPromptPart,
 )
+from ..settings import ModelSettings
+from ..tools import ToolDefinition
+from . import AgentModel, EitherStreamedResponse, Model, StreamStructuredResponse, StreamTextResponse
 
 
 @dataclass(init=False)
@@ -59,13 +66,13 @@ class FunctionModel(Model):
 
     async def agent_model(
         self,
-        function_tools: Mapping[str, AbstractToolDefinition],
+        *,
+        function_tools: list[ToolDefinition],
         allow_text_result: bool,
-        result_tools: Sequence[AbstractToolDefinition] | None,
+        result_tools: list[ToolDefinition],
     ) -> AgentModel:
-        result_tools = list(result_tools) if result_tools is not None else None
         return FunctionAgentModel(
-            self.function, self.stream_function, AgentInfo(function_tools, allow_text_result, result_tools)
+            self.function, self.stream_function, AgentInfo(function_tools, allow_text_result, result_tools, None)
         )
 
     def name(self) -> str:
@@ -84,7 +91,7 @@ class AgentInfo:
     This is passed as the second to functions used within [`FunctionModel`][pydantic_ai.models.function.FunctionModel].
     """
 
-    function_tools: Mapping[str, AbstractToolDefinition]
+    function_tools: list[ToolDefinition]
     """The function tools available on this agent.
 
     These are the tools registered via the [`tool`][pydantic_ai.Agent.tool] and
@@ -92,8 +99,10 @@ class AgentInfo:
     """
     allow_text_result: bool
     """Whether a plain text result is allowed."""
-    result_tools: list[AbstractToolDefinition] | None
+    result_tools: list[ToolDefinition]
     """The tools that can called as the final result of the run."""
+    model_settings: ModelSettings | None
+    """The model settings passed to the run call."""
 
 
 @dataclass
@@ -112,10 +121,10 @@ class DeltaToolCall:
 DeltaToolCalls: TypeAlias = dict[int, DeltaToolCall]
 """A mapping of tool call IDs to incremental changes."""
 
-FunctionDef: TypeAlias = Callable[[list[Message], AgentInfo], Union[ModelAnyResponse, Awaitable[ModelAnyResponse]]]
+FunctionDef: TypeAlias = Callable[[list[ModelMessage], AgentInfo], Union[ModelResponse, Awaitable[ModelResponse]]]
 """A function used to generate a non-streamed response."""
 
-StreamFunctionDef: TypeAlias = Callable[[list[Message], AgentInfo], AsyncIterator[Union[str, DeltaToolCalls]]]
+StreamFunctionDef: TypeAlias = Callable[[list[ModelMessage], AgentInfo], AsyncIterator[Union[str, DeltaToolCalls]]]
 """A function used to generate a streamed response.
 
 While this is defined as having return type of `AsyncIterator[Union[str, DeltaToolCalls]]`, it should
@@ -133,18 +142,25 @@ class FunctionAgentModel(AgentModel):
     stream_function: StreamFunctionDef | None
     agent_info: AgentInfo
 
-    async def request(self, messages: list[Message]) -> tuple[ModelAnyResponse, result.Cost]:
+    async def request(
+        self, messages: list[ModelMessage], model_settings: ModelSettings | None
+    ) -> tuple[ModelResponse, result.Cost]:
+        agent_info = replace(self.agent_info, model_settings=model_settings)
+
         assert self.function is not None, 'FunctionModel must receive a `function` to support non-streamed requests'
         if inspect.iscoroutinefunction(self.function):
-            response = await self.function(messages, self.agent_info)
+            response = await self.function(messages, agent_info)
         else:
-            response_ = await _utils.run_in_executor(self.function, messages, self.agent_info)
-            response = cast(ModelAnyResponse, response_)
+            response_ = await _utils.run_in_executor(self.function, messages, agent_info)
+            assert isinstance(response_, ModelResponse), response_
+            response = response_
         # TODO is `messages` right here? Should it just be new messages?
         return response, _estimate_cost(chain(messages, [response]))
 
     @asynccontextmanager
-    async def request_stream(self, messages: list[Message]) -> AsyncIterator[EitherStreamedResponse]:
+    async def request_stream(
+        self, messages: list[ModelMessage], model_settings: ModelSettings | None
+    ) -> AsyncIterator[EitherStreamedResponse]:
         assert (
             self.stream_function is not None
         ), 'FunctionModel must receive a `stream_function` to support streamed requests'
@@ -212,13 +228,13 @@ class FunctionStreamStructuredResponse(StreamStructuredResponse):
             else:
                 self._delta_tool_calls[key] = new
 
-    def get(self, *, final: bool = False) -> ModelStructuredResponse:
-        calls: list[ToolCall] = []
+    def get(self, *, final: bool = False) -> ModelResponse:
+        calls: list[ModelResponsePart] = []
         for c in self._delta_tool_calls.values():
             if c.name is not None and c.json_args is not None:
-                calls.append(ToolCall.from_json(c.name, c.json_args))
+                calls.append(ToolCallPart.from_json(c.name, c.json_args))
 
-        return ModelStructuredResponse(calls, timestamp=self._timestamp)
+        return ModelResponse(calls, timestamp=self._timestamp)
 
     def cost(self) -> result.Cost:
         return result.Cost()
@@ -227,32 +243,38 @@ class FunctionStreamStructuredResponse(StreamStructuredResponse):
         return self._timestamp
 
 
-def _estimate_cost(messages: Iterable[Message]) -> result.Cost:
+def _estimate_cost(messages: Iterable[ModelMessage]) -> result.Cost:
     """Very rough guesstimate of the number of tokens associate with a series of messages.
 
     This is designed to be used solely to give plausible numbers for testing!
     """
     # there seem to be about 50 tokens of overhead for both Gemini and OpenAI calls, so add that here ¯\_(ツ)_/¯
-
     request_tokens = 50
     response_tokens = 0
     for message in messages:
-        if message.role == 'system' or message.role == 'user':
-            request_tokens += _string_cost(message.content)
-        elif message.role == 'tool-return':
-            request_tokens += _string_cost(message.model_response_str())
-        elif message.role == 'retry-prompt':
-            request_tokens += _string_cost(message.model_response())
-        elif message.role == 'model-text-response':
-            response_tokens += _string_cost(message.content)
-        elif message.role == 'model-structured-response':
-            for call in message.calls:
-                if isinstance(call.args, ArgsJson):
-                    args_str = call.args.args_json
+        if isinstance(message, ModelRequest):
+            for part in message.parts:
+                if isinstance(part, (SystemPromptPart, UserPromptPart)):
+                    request_tokens += _string_cost(part.content)
+                elif isinstance(part, ToolReturnPart):
+                    request_tokens += _string_cost(part.model_response_str())
+                elif isinstance(part, RetryPromptPart):
+                    request_tokens += _string_cost(part.model_response())
                 else:
-                    args_str = pydantic_core.to_json(call.args.args_dict).decode()
-
-                response_tokens += 1 + _string_cost(args_str)
+                    assert_never(part)
+        elif isinstance(message, ModelResponse):
+            for part in message.parts:
+                if isinstance(part, TextPart):
+                    response_tokens += _string_cost(part.content)
+                elif isinstance(part, ToolCallPart):
+                    call = part
+                    if isinstance(call.args, ArgsJson):
+                        args_str = call.args.args_json
+                    else:
+                        args_str = pydantic_core.to_json(call.args.args_dict).decode()
+                    response_tokens += 1 + _string_cost(args_str)
+                else:
+                    assert_never(part)
         else:
             assert_never(message)
     return result.Cost(