pydantic-ai-slim 1.0.10__py3-none-any.whl → 1.0.12__py3-none-any.whl

pydantic_ai/messages.py

@@ -114,6 +114,20 @@ class FileUrl(ABC):
 
     _: KW_ONLY
 
+    identifier: str
+    """The identifier of the file, such as a unique ID. generating one from the url if not explicitly set.
+
+    This identifier can be provided to the model in a message to allow it to refer to this file in a tool call argument,
+    and the tool can look up the file in question by iterating over the message history and finding the matching `FileUrl`.
+
+    This identifier is only automatically passed to the model when the `FileUrl` is returned by a tool.
+    If you're passing the `FileUrl` as a user message, it's up to you to include a separate text part with the identifier,
+    e.g. "This is file <identifier>:" preceding the `FileUrl`.
+
+    It's also included in inline-text delimiters for providers that require inlining text documents, so the model can
+    distinguish multiple files.
+    """
+
     force_download: bool = False
     """If the model supports it:
 
@@ -126,23 +140,13 @@ class FileUrl(ABC):
 
     Supported by:
     - `GoogleModel`: `VideoUrl.vendor_metadata` is used as `video_metadata`: https://ai.google.dev/gemini-api/docs/video-understanding#customize-video-processing
+    - `OpenAIChatModel`, `OpenAIResponsesModel`: `ImageUrl.vendor_metadata['detail']` is used as `detail` setting for images
     """
 
     _media_type: Annotated[str | None, pydantic.Field(alias='media_type', default=None, exclude=True)] = field(
         compare=False, default=None
     )
 
-    identifier: str | None = None
-    """The identifier of the file, such as a unique ID. generating one from the url if not explicitly set
-
-    This identifier can be provided to the model in a message to allow it to refer to this file in a tool call argument,
-    and the tool can look up the file in question by iterating over the message history and finding the matching `FileUrl`.
-
-    This identifier is only automatically passed to the model when the `FileUrl` is returned by a tool.
-    If you're passing the `FileUrl` as a user message, it's up to you to include a separate text part with the identifier,
-    e.g. "This is file <identifier>:" preceding the `FileUrl`.
-    """
-
     def __init__(
         self,
         url: str,
@@ -471,6 +475,7 @@ class BinaryContent:
 
     Supported by:
     - `GoogleModel`: `BinaryContent.vendor_metadata` is used as `video_metadata`: https://ai.google.dev/gemini-api/docs/video-understanding#customize-video-processing
+    - `OpenAIChatModel`, `OpenAIResponsesModel`: `BinaryContent.vendor_metadata['detail']` is used as `detail` setting for images
     """
 
     kind: Literal['binary'] = 'binary'

pydantic_ai/models/__init__.py

@@ -20,6 +20,7 @@ import httpx
 from typing_extensions import TypeAliasType, TypedDict
 
 from .. import _utils
+from .._json_schema import JsonSchemaTransformer
 from .._output import OutputObjectDefinition
 from .._parts_manager import ModelResponsePartsManager
 from .._run_context import RunContext
@@ -40,7 +41,6 @@ from ..messages import (
 )
 from ..output import OutputMode
 from ..profiles import DEFAULT_PROFILE, ModelProfile, ModelProfileSpec
-from ..profiles._json_schema import JsonSchemaTransformer
 from ..settings import ModelSettings
 from ..tools import ToolDefinition
 from ..usage import RequestUsage
@@ -65,6 +65,8 @@ KnownModelName = TypeAliasType(
         'anthropic:claude-opus-4-20250514',
         'anthropic:claude-sonnet-4-0',
         'anthropic:claude-sonnet-4-20250514',
+        'anthropic:claude-sonnet-4-5',
+        'anthropic:claude-sonnet-4-5-20250929',
         'bedrock:amazon.titan-tg1-large',
         'bedrock:amazon.titan-text-lite-v1',
         'bedrock:amazon.titan-text-express-v1',
@@ -121,23 +123,6 @@ KnownModelName = TypeAliasType(
         'cerebras:qwen-3-32b',
         'cerebras:qwen-3-coder-480b',
         'cerebras:qwen-3-235b-a22b-thinking-2507',
-        'claude-3-5-haiku-20241022',
-        'claude-3-5-haiku-latest',
-        'claude-3-5-sonnet-20240620',
-        'claude-3-5-sonnet-20241022',
-        'claude-3-5-sonnet-latest',
-        'claude-3-7-sonnet-20250219',
-        'claude-3-7-sonnet-latest',
-        'claude-3-haiku-20240307',
-        'claude-3-opus-20240229',
-        'claude-3-opus-latest',
-        'claude-4-opus-20250514',
-        'claude-4-sonnet-20250514',
-        'claude-opus-4-0',
-        'claude-opus-4-1-20250805',
-        'claude-opus-4-20250514',
-        'claude-sonnet-4-0',
-        'claude-sonnet-4-20250514',
         'cohere:c4ai-aya-expanse-32b',
         'cohere:c4ai-aya-expanse-8b',
         'cohere:command',
@@ -163,54 +148,6 @@ KnownModelName = TypeAliasType(
         'google-vertex:gemini-2.5-flash',
         'google-vertex:gemini-2.5-flash-lite',
         'google-vertex:gemini-2.5-pro',
-        'gpt-3.5-turbo',
-        'gpt-3.5-turbo-0125',
-        'gpt-3.5-turbo-0301',
-        'gpt-3.5-turbo-0613',
-        'gpt-3.5-turbo-1106',
-        'gpt-3.5-turbo-16k',
-        'gpt-3.5-turbo-16k-0613',
-        'gpt-4',
-        'gpt-4-0125-preview',
-        'gpt-4-0314',
-        'gpt-4-0613',
-        'gpt-4-1106-preview',
-        'gpt-4-32k',
-        'gpt-4-32k-0314',
-        'gpt-4-32k-0613',
-        'gpt-4-turbo',
-        'gpt-4-turbo-2024-04-09',
-        'gpt-4-turbo-preview',
-        'gpt-4-vision-preview',
-        'gpt-4.1',
-        'gpt-4.1-2025-04-14',
-        'gpt-4.1-mini',
-        'gpt-4.1-mini-2025-04-14',
-        'gpt-4.1-nano',
-        'gpt-4.1-nano-2025-04-14',
-        'gpt-4o',
-        'gpt-4o-2024-05-13',
-        'gpt-4o-2024-08-06',
-        'gpt-4o-2024-11-20',
-        'gpt-4o-audio-preview',
-        'gpt-4o-audio-preview-2024-10-01',
-        'gpt-4o-audio-preview-2024-12-17',
-        'gpt-4o-audio-preview-2025-06-03',
-        'gpt-4o-mini',
-        'gpt-4o-mini-2024-07-18',
-        'gpt-4o-mini-audio-preview',
-        'gpt-4o-mini-audio-preview-2024-12-17',
-        'gpt-4o-mini-search-preview',
-        'gpt-4o-mini-search-preview-2025-03-11',
-        'gpt-4o-search-preview',
-        'gpt-4o-search-preview-2025-03-11',
-        'gpt-5',
-        'gpt-5-2025-08-07',
-        'gpt-5-chat-latest',
-        'gpt-5-mini',
-        'gpt-5-mini-2025-08-07',
-        'gpt-5-nano',
-        'gpt-5-nano-2025-08-07',
         'grok:grok-4',
         'grok:grok-4-0709',
         'grok:grok-3',
@@ -271,22 +208,6 @@ KnownModelName = TypeAliasType(
         'moonshotai:kimi-latest',
         'moonshotai:kimi-thinking-preview',
         'moonshotai:kimi-k2-0711-preview',
-        'o1',
-        'o1-2024-12-17',
-        'o1-mini',
-        'o1-mini-2024-09-12',
-        'o1-preview',
-        'o1-preview-2024-09-12',
-        'o1-pro',
-        'o1-pro-2025-03-19',
-        'o3',
-        'o3-2025-04-16',
-        'o3-deep-research',
-        'o3-deep-research-2025-06-26',
-        'o3-mini',
-        'o3-mini-2025-01-31',
-        'o3-pro',
-        'o3-pro-2025-06-10',
         'openai:chatgpt-4o-latest',
         'openai:codex-mini-latest',
         'openai:gpt-3.5-turbo',

pydantic_ai/models/anthropic.py

@@ -10,11 +10,10 @@ from typing import Any, Literal, cast, overload
 from pydantic import TypeAdapter
 from typing_extensions import assert_never
 
-from pydantic_ai.builtin_tools import CodeExecutionTool, WebSearchTool
-
 from .. import ModelHTTPError, UnexpectedModelBehavior, _utils, usage
 from .._run_context import RunContext
 from .._utils import guard_tool_call_id as _guard_tool_call_id
+from ..builtin_tools import CodeExecutionTool, MemoryTool, WebSearchTool
 from ..exceptions import UserError
 from ..messages import (
     BinaryContent,
@@ -54,7 +53,7 @@ _FINISH_REASON_MAP: dict[BetaStopReason, FinishReason] = {
 
 
 try:
-    from anthropic import NOT_GIVEN, APIStatusError, AsyncStream
+    from anthropic import NOT_GIVEN, APIStatusError, AsyncStream, omit as OMIT
     from anthropic.types.beta import (
         BetaBase64PDFBlockParam,
         BetaBase64PDFSourceParam,
@@ -68,6 +67,7 @@ try:
         BetaContentBlockParam,
         BetaImageBlockParam,
         BetaInputJSONDelta,
+        BetaMemoryTool20250818Param,
         BetaMessage,
         BetaMessageParam,
         BetaMetadataParam,
@@ -255,8 +255,7 @@ class AnthropicModel(Model):
     ) -> BetaMessage | AsyncStream[BetaRawMessageStreamEvent]:
         # standalone function to make it easier to override
         tools = self._get_tools(model_request_parameters)
-        builtin_tools, tool_headers = self._get_builtin_tools(model_request_parameters)
-        tools += builtin_tools
+        tools, beta_features = self._add_builtin_tools(tools, model_request_parameters)
 
         tool_choice: BetaToolChoiceParam | None
 
@@ -265,6 +264,10 @@ class AnthropicModel(Model):
         else:
             if not model_request_parameters.allow_text_output:
                 tool_choice = {'type': 'any'}
+                if (thinking := model_settings.get('anthropic_thinking')) and thinking.get('type') == 'enabled':
+                    raise UserError(
+                        'Anthropic does not support thinking and output tools at the same time. Use `output_type=PromptedOutput(...)` instead.'
+                    )
             else:
                 tool_choice = {'type': 'auto'}
 
@@ -275,24 +278,26 @@ class AnthropicModel(Model):
 
         try:
             extra_headers = model_settings.get('extra_headers', {})
-            for k, v in tool_headers.items():
-                extra_headers.setdefault(k, v)
             extra_headers.setdefault('User-Agent', get_user_agent())
+            if beta_features:
+                if 'anthropic-beta' in extra_headers:
+                    beta_features.insert(0, extra_headers['anthropic-beta'])
+                extra_headers['anthropic-beta'] = ','.join(beta_features)
 
             return await self.client.beta.messages.create(
                 max_tokens=model_settings.get('max_tokens', 4096),
-                system=system_prompt or NOT_GIVEN,
+                system=system_prompt or OMIT,
                 messages=anthropic_messages,
                 model=self._model_name,
-                tools=tools or NOT_GIVEN,
-                tool_choice=tool_choice or NOT_GIVEN,
+                tools=tools or OMIT,
+                tool_choice=tool_choice or OMIT,
                 stream=stream,
-                thinking=model_settings.get('anthropic_thinking', NOT_GIVEN),
-                stop_sequences=model_settings.get('stop_sequences', NOT_GIVEN),
-                temperature=model_settings.get('temperature', NOT_GIVEN),
-                top_p=model_settings.get('top_p', NOT_GIVEN),
+                thinking=model_settings.get('anthropic_thinking', OMIT),
+                stop_sequences=model_settings.get('stop_sequences', OMIT),
+                temperature=model_settings.get('temperature', OMIT),
+                top_p=model_settings.get('top_p', OMIT),
                 timeout=model_settings.get('timeout', NOT_GIVEN),
-                metadata=model_settings.get('anthropic_metadata', NOT_GIVEN),
+                metadata=model_settings.get('anthropic_metadata', OMIT),
                 extra_headers=extra_headers,
                 extra_body=model_settings.get('extra_body'),
             )
@@ -363,14 +368,13 @@ class AnthropicModel(Model):
             _provider_name=self._provider.name,
         )
 
-    def _get_tools(self, model_request_parameters: ModelRequestParameters) -> list[BetaToolParam]:
+    def _get_tools(self, model_request_parameters: ModelRequestParameters) -> list[BetaToolUnionParam]:
         return [self._map_tool_definition(r) for r in model_request_parameters.tool_defs.values()]
 
-    def _get_builtin_tools(
-        self, model_request_parameters: ModelRequestParameters
-    ) -> tuple[list[BetaToolUnionParam], dict[str, str]]:
-        tools: list[BetaToolUnionParam] = []
-        extra_headers: dict[str, str] = {}
+    def _add_builtin_tools(
+        self, tools: list[BetaToolUnionParam], model_request_parameters: ModelRequestParameters
+    ) -> tuple[list[BetaToolUnionParam], list[str]]:
+        beta_features: list[str] = []
         for tool in model_request_parameters.builtin_tools:
             if isinstance(tool, WebSearchTool):
                 user_location = UserLocation(type='approximate', **tool.user_location) if tool.user_location else None
@@ -385,13 +389,20 @@ class AnthropicModel(Model):
                     )
                 )
             elif isinstance(tool, CodeExecutionTool):  # pragma: no branch
-                extra_headers['anthropic-beta'] = 'code-execution-2025-05-22'
                 tools.append(BetaCodeExecutionTool20250522Param(name='code_execution', type='code_execution_20250522'))
+                beta_features.append('code-execution-2025-05-22')
+            elif isinstance(tool, MemoryTool):  # pragma: no branch
+                if 'memory' not in model_request_parameters.tool_defs:
+                    raise UserError("Built-in `MemoryTool` requires a 'memory' tool to be defined.")
+                # Replace the memory tool definition with the built-in memory tool
+                tools = [tool for tool in tools if tool['name'] != 'memory']
+                tools.append(BetaMemoryTool20250818Param(name='memory', type='memory_20250818'))
+                beta_features.append('context-management-2025-06-27')
             else:  # pragma: no cover
                 raise UserError(
                     f'`{tool.__class__.__name__}` is not supported by `AnthropicModel`. If it should be, please file an issue.'
                 )
-        return tools, extra_headers
+        return tools, beta_features
 
     async def _map_message(self, messages: list[ModelMessage]) -> tuple[str, list[BetaMessageParam]]:  # noqa: C901
         """Just maps a `pydantic_ai.Message` to a `anthropic.types.MessageParam`."""
@@ -759,6 +770,8 @@ def _map_server_tool_use_block(item: BetaServerToolUseBlock, provider_name: str)
             args=cast(dict[str, Any], item.input) or None,
             tool_call_id=item.id,
         )
+    elif item.name in ('web_fetch', 'bash_code_execution', 'text_editor_code_execution'):  # pragma: no cover
+        raise NotImplementedError(f'Anthropic built-in tool {item.name!r} is not currently supported.')
     else:
         assert_never(item.name)
 

pydantic_ai/models/bedrock.py

@@ -13,10 +13,7 @@ import anyio
 import anyio.to_thread
 from typing_extensions import ParamSpec, assert_never
 
-from pydantic_ai import _utils, usage
-from pydantic_ai._run_context import RunContext
-from pydantic_ai.exceptions import UserError
-from pydantic_ai.messages import (
+from pydantic_ai import (
     AudioUrl,
     BinaryContent,
     BuiltinToolCallPart,
@@ -25,6 +22,7 @@ from pydantic_ai.messages import (
     FinishReason,
     ImageUrl,
     ModelMessage,
+    ModelProfileSpec,
     ModelRequest,
     ModelResponse,
     ModelResponsePart,
@@ -37,9 +35,12 @@ from pydantic_ai.messages import (
     ToolReturnPart,
     UserPromptPart,
     VideoUrl,
+    _utils,
+    usage,
 )
+from pydantic_ai._run_context import RunContext
+from pydantic_ai.exceptions import UserError
 from pydantic_ai.models import Model, ModelRequestParameters, StreamedResponse, download_item
-from pydantic_ai.profiles import ModelProfileSpec
 from pydantic_ai.providers import Provider, infer_provider
 from pydantic_ai.providers.bedrock import BedrockModelProfile
 from pydantic_ai.settings import ModelSettings

pydantic_ai/models/google.py

@@ -419,8 +419,8 @@ class GoogleModel(Model):
         return contents, config
 
     def _process_response(self, response: GenerateContentResponse) -> ModelResponse:
-        if not response.candidates or len(response.candidates) != 1:
-            raise UnexpectedModelBehavior('Expected exactly one candidate in Gemini response')  # pragma: no cover
+        if not response.candidates:
+            raise UnexpectedModelBehavior('Expected at least one candidate in Gemini response')  # pragma: no cover
         candidate = response.candidates[0]
         if candidate.content is None or candidate.content.parts is None:
             if candidate.finish_reason == 'SAFETY':

pydantic_ai/models/instrumented.py

@@ -9,6 +9,7 @@ from dataclasses import dataclass, field
 from typing import Any, Literal, cast
 from urllib.parse import urlparse
 
+from genai_prices.types import PriceCalculation
 from opentelemetry._events import (
     Event,  # pyright: ignore[reportPrivateImportUsage]
     EventLogger,  # pyright: ignore[reportPrivateImportUsage]
@@ -169,6 +170,11 @@ class InstrumentationSettings:
             self.tokens_histogram = self.meter.create_histogram(
                 **tokens_histogram_kwargs,  # pyright: ignore
             )
+        self.cost_histogram = self.meter.create_histogram(
+            'operation.cost',
+            unit='{USD}',
+            description='Monetary cost',
+        )
 
     def messages_to_otel_events(self, messages: list[ModelMessage]) -> list[Event]:
         """Convert a list of model messages to OpenTelemetry events.
@@ -302,6 +308,21 @@ class InstrumentationSettings:
                 }
             )
 
+    def record_metrics(
+        self,
+        response: ModelResponse,
+        price_calculation: PriceCalculation | None,
+        attributes: dict[str, AttributeValue],
+    ):
+        for typ in ['input', 'output']:
+            if not (tokens := getattr(response.usage, f'{typ}_tokens', 0)):  # pragma: no cover
+                continue
+            token_attributes = {**attributes, 'gen_ai.token.type': typ}
+            self.tokens_histogram.record(tokens, token_attributes)
+            if price_calculation:
+                cost = float(getattr(price_calculation, f'{typ}_price'))
+                self.cost_histogram.record(cost, token_attributes)
+
 
 GEN_AI_SYSTEM_ATTRIBUTE = 'gen_ai.system'
 GEN_AI_REQUEST_MODEL_ATTRIBUTE = 'gen_ai.request.model'
@@ -395,6 +416,7 @@ class InstrumentedModel(WrapperModel):
                     system = cast(str, attributes[GEN_AI_SYSTEM_ATTRIBUTE])
 
                     response_model = response.model_name or request_model
+                    price_calculation = None
 
                     def _record_metrics():
                         metric_attributes = {
@@ -403,16 +425,7 @@ class InstrumentedModel(WrapperModel):
                             'gen_ai.request.model': request_model,
                             'gen_ai.response.model': response_model,
                         }
-                        if response.usage.input_tokens:  # pragma: no branch
-                            self.instrumentation_settings.tokens_histogram.record(
-                                response.usage.input_tokens,
-                                {**metric_attributes, 'gen_ai.token.type': 'input'},
-                            )
-                        if response.usage.output_tokens:  # pragma: no branch
-                            self.instrumentation_settings.tokens_histogram.record(
-                                response.usage.output_tokens,
-                                {**metric_attributes, 'gen_ai.token.type': 'output'},
-                            )
+                        self.instrumentation_settings.record_metrics(response, price_calculation, metric_attributes)
 
                     nonlocal record_metrics
                     record_metrics = _record_metrics
@@ -427,7 +440,7 @@ class InstrumentedModel(WrapperModel):
                         'gen_ai.response.model': response_model,
                     }
                     try:
-                        attributes_to_set['operation.cost'] = float(response.cost().total_price)
+                        price_calculation = response.cost()
                     except LookupError:
                         # The cost of this provider/model is unknown, which is common.
                         pass
@@ -435,6 +448,9 @@ class InstrumentedModel(WrapperModel):
                         warnings.warn(
                             f'Failed to get cost from response: {type(e).__name__}: {e}', CostCalculationFailedWarning
                         )
+                    else:
+                        attributes_to_set['operation.cost'] = float(price_calculation.total_price)
+
                     if response.provider_response_id is not None:
                         attributes_to_set['gen_ai.response.id'] = response.provider_response_id
                     if response.finish_reason is not None: