instructor API reference

183 public APIs from instructor (instructor-ai/instructor) — 74 classes, 39 functions, 70 methods. Signatures extracted by static analysis of the actual source.

Repository: instructor-ai/instructor

Kind	Count
Classes	74
Functions	39
Methods	70

API list

classinstructor.batch.BatchJob

Legacy BatchJob class for backward compatibility

classinstructor.batch.models.BatchError

Error information for failed batch requests

classinstructor.batch.models.BatchErrorInfo

Batch-level error information

classinstructor.batch.models.BatchFiles

File references for batch job

classinstructor.batch.models.BatchRequestCounts

Unified request counts across providers

classinstructor.batch.models.BatchStatus

Normalized batch status across providers

classinstructor.batch.models.BatchSuccess

Successful batch result with custom_id

classinstructor.batch.models.BatchTimestamps

Comprehensive timestamp tracking

classinstructor.batch.processor.BatchProcessor

Unified batch processor that works across all providers

methodinstructor.batch.processor.BatchProcessor.get_batch_status(batch_id:str) -> dict[str, Any]

Get batch job status from the provider

methodinstructor.batch.processor.BatchProcessor.retrieve_results(batch_id:str) -> list[BatchResult]

Retrieve and parse batch results from the provider

classinstructor.batch.providers.anthropic.AnthropicProvider

Anthropic batch processing provider

methodinstructor.batch.providers.anthropic.AnthropicProvider.cancel_batch(batch_id:str) -> dict[str, Any]

Cancel Anthropic batch job

methodinstructor.batch.providers.anthropic.AnthropicProvider.delete_batch(batch_id:str) -> dict[str, Any]

Delete Anthropic batch job

methodinstructor.batch.providers.anthropic.AnthropicProvider.download_results(batch_id:str, file_path:str) -> None

Download Anthropic batch results to a file

methodinstructor.batch.providers.anthropic.AnthropicProvider.get_status(batch_id:str) -> dict[str, Any]

Get Anthropic batch status

methodinstructor.batch.providers.anthropic.AnthropicProvider.list_batches(limit:int=10) -> list[BatchJobInfo]

List Anthropic batch jobs

methodinstructor.batch.providers.anthropic.AnthropicProvider.retrieve_results(batch_id:str) -> str

Retrieve Anthropic batch results

method

instructor.batch.providers.anthropic.AnthropicProvider.submit_batch(file_path_or_buffer:Union[str, io.BytesIO], metadata:Optional[dict[str, Any]]=None, **kwargs) -> str

Submit Anthropic batch job

classinstructor.batch.providers.base.BatchProvider

Abstract base class for batch processing providers

methodinstructor.batch.providers.base.BatchProvider.cancel_batch(batch_id:str) -> dict[str, Any]

Cancel a batch job

methodinstructor.batch.providers.base.BatchProvider.delete_batch(batch_id:str) -> dict[str, Any]

Delete a batch job

methodinstructor.batch.providers.base.BatchProvider.download_results(batch_id:str, file_path:str) -> None

Download batch results to a file

methodinstructor.batch.providers.base.BatchProvider.get_status(batch_id:str) -> dict[str, Any]

Get the status of a batch job

methodinstructor.batch.providers.base.BatchProvider.list_batches(limit:int=10) -> list[BatchJobInfo]

List batch jobs

methodinstructor.batch.providers.base.BatchProvider.retrieve_results(batch_id:str) -> str

Retrieve batch results as a string

method

instructor.batch.providers.base.BatchProvider.submit_batch(file_path_or_buffer:Union[str, io.BytesIO], metadata:Optional[dict[str, Any]]=None, **kwargs) -> str

Submit a batch job and return the job ID

funcinstructor.batch.providers.get_provider(provider_name:str) -> BatchProvider

Factory function to get the appropriate provider instance

classinstructor.batch.providers.openai.OpenAIProvider

OpenAI batch processing provider

methodinstructor.batch.providers.openai.OpenAIProvider.cancel_batch(batch_id:str) -> dict[str, Any]

Cancel OpenAI batch job

methodinstructor.batch.providers.openai.OpenAIProvider.delete_batch(batch_id:str) -> dict[str, Any]

Delete OpenAI batch job

methodinstructor.batch.providers.openai.OpenAIProvider.download_results(batch_id:str, file_path:str) -> None

Download OpenAI batch results to a file

methodinstructor.batch.providers.openai.OpenAIProvider.get_status(batch_id:str) -> dict[str, Any]

Get OpenAI batch status

methodinstructor.batch.providers.openai.OpenAIProvider.list_batches(limit:int=10) -> list[BatchJobInfo]

List OpenAI batch jobs

methodinstructor.batch.providers.openai.OpenAIProvider.retrieve_results(batch_id:str) -> str

Retrieve OpenAI batch results

method

instructor.batch.providers.openai.OpenAIProvider.submit_batch(file_path_or_buffer:Union[str, io.BytesIO], metadata:Optional[dict[str, Any]]=None, **kwargs) -> str

Submit OpenAI batch job

methodinstructor.batch.request.BatchRequest.get_json_schema() -> dict[str, Any]

Generate JSON schema from response_model

funcinstructor.batch.utils.extract_results(results:list[BatchResult]) -> list[T]

Extract just the result objects from successful results

funcinstructor.batch.utils.filter_errors(results:list[BatchResult]) -> list[BatchError]

Filter to only error results

funcinstructor.batch.utils.filter_successful(results:list[BatchResult]) -> list[BatchSuccess[T]]

Filter to only successful results

funcinstructor.batch.utils.get_results_by_custom_id(results:list[BatchResult]) -> dict[str, BatchResult]

Create a dictionary mapping custom_id to results

classinstructor.cache.BaseCache

Abstract cache contract.

methodinstructor.cache.BaseCache.get(key:str) -> Any | None

Return *None* to indicate a cache miss.

methodinstructor.cache.BaseCache.set(key:str, value:Any, ttl:int | None=None) -> None

Store *value* under *key*.

classinstructor.cache.DiskCache

Wrapper around `diskcache.Cache`.

func

instructor.cache.make_cache_key(*messages:Any, *model:str | None, *response_model:type[BaseModel] | None, *mode:str | None=None) -> str

Compute a *deterministic* cache key.

funcinstructor.cli.cli.docs(query:Optional[str]=typer.Argument(None, help='Search the documentation')) -> None

Open the instructor documentation website.

funcinstructor.cli.deprecated_hub.hub() -> None

This command has been deprecated.

funcinstructor.cli.usage.get_model_cost(model:ModelNames) -> Union[dict[str, float], float]

Get the cost details for a given model.

funcinstructor.distil.format_function(func:Callable[..., Any]) -> str

Format a function as a string with docstring and body.

funcinstructor.distil.get_signature_from_fn(fn:Callable[..., Any]) -> str

Get the function signature as a string.

classinstructor.v2.core.client.Instructor

Sync client wrapper that adds structured output support.

methodinstructor.v2.core.client.Instructor.handle_kwargs(kwargs:dict[str, Any]) -> dict[str, Any]

Handle and process keyword arguments for the API call.

classinstructor.v2.core.client.Response

Helper for responses API using a patched client.

funcinstructor.v2.core.decorators.decorator(handler_class:type) -> type

classinstructor.v2.core.errors.AsyncValidationError

Exception raised during async validation.

classinstructor.v2.core.errors.ConfigurationError

Exception raised for configuration-related errors.

classinstructor.v2.core.errors.FailedAttempt

Represents a single failed retry attempt.

classinstructor.v2.core.errors.InstructorError

Base exception for all Instructor-specific errors.

classinstructor.v2.core.errors.MultimodalError

Exception raised for multimodal content processing errors.

classinstructor.v2.core.errors.ProviderError

Exception raised for provider-specific errors.

classinstructor.v2.core.errors.ResponseParsingError

Exception raised when unable to parse the LLM response.

classinstructor.v2.core.errors.ValidationError

Exception raised when LLM response validation fails.

classinstructor.v2.core.exceptions.RegistryValidationMixin

Mixin providing registry validation helper methods.

funcinstructor.v2.core.function_calls.response_schema(cls:type[Model]) -> type[Model]

Wrap a Pydantic model class to add ResponseSchema behavior.

classinstructor.v2.core.handler.ModeHandler

Base class for mode handlers.

classinstructor.v2.core.hooks.CompletionKwargsHandler

Protocol for completion kwargs handlers.

classinstructor.v2.core.hooks.CompletionResponseHandler

Protocol for completion response handlers.

classinstructor.v2.core.hooks.Hooks

Hooks class for handling and emitting events related to completion processes.

methodinstructor.v2.core.hooks.Hooks.clear(hook_name:HookNameType | None=None) -> None

Clear handlers for a specific event or all events.

methodinstructor.v2.core.hooks.Hooks.combine(*hooks_instances:Hooks) -> Hooks

Combine multiple Hooks instances into a new one.

methodinstructor.v2.core.hooks.Hooks.copy() -> Hooks

Create a deep copy of this Hooks instance.

methodinstructor.v2.core.hooks.Hooks.emit(hook_name:HookName, *args:Any, **kwargs:Any) -> None

Generic method to emit events for any hook type.

methodinstructor.v2.core.hooks.Hooks.emit_completion_arguments(*args:Any, **kwargs:Any) -> None

Emit a completion arguments event.

methodinstructor.v2.core.hooks.Hooks.emit_completion_error(error:Exception, **kwargs:Any) -> None

Emit a completion error event.

methodinstructor.v2.core.hooks.Hooks.emit_completion_last_attempt(error:Exception, **kwargs:Any) -> None

Emit a completion last attempt event.

methodinstructor.v2.core.hooks.Hooks.emit_completion_response(response:Any) -> None

Emit a completion response event.

methodinstructor.v2.core.hooks.Hooks.emit_parse_error(error:Exception, **kwargs:Any) -> None

Emit a parse error event.

methodinstructor.v2.core.hooks.Hooks.get_hook_name(hook_name:HookNameType) -> HookName

Convert a string hook name to its corresponding enum value.

methodinstructor.v2.core.hooks.Hooks.off(hook_name:HookNameType, handler:HandlerType) -> None

Remove a specific handler from an event.

methodinstructor.v2.core.hooks.Hooks.on(hook_name:HookNameType, handler:HandlerType) -> None

classinstructor.v2.core.hooks.ParseErrorHandler

Protocol for parse error handlers.

classinstructor.v2.core.mode.Mode

Mode enumeration for patching LLM API clients.

methodinstructor.v2.core.mode.Mode.json_modes() -> set['Mode']

Returns a set of all JSON-based modes.

methodinstructor.v2.core.mode.Mode.tool_modes() -> set['Mode']

Returns a set of all tool-based modes.

methodinstructor.v2.core.mode.Mode.warn_deprecated_mode(mode:'Mode') -> None

Warn about provider-specific mode deprecation.

funcinstructor.v2.core.mode.reset_deprecated_mode_warnings() -> None

Reset deprecation warning tracking.

methodinstructor.v2.core.multimodal.Audio.autodetect(source:str | Path) -> Audio

Attempt to autodetect an audio from a source string or Path.

methodinstructor.v2.core.multimodal.Audio.from_gs_url(data_uri:str, timeout:int=30) -> Audio

Create an Audio instance from a Google Cloud Storage URL.

methodinstructor.v2.core.multimodal.Audio.from_path(path:Union[str, Path]) -> Audio

Create an Audio instance from a file path.

methodinstructor.v2.core.multimodal.Audio.from_url(url:str) -> Audio

Create an Audio instance from a URL.

methodinstructor.v2.core.multimodal.Image.from_gs_url(data_uri:str, timeout:int=30) -> Image

Create an Image instance from a Google Cloud Storage URL.

classinstructor.v2.core.multimodal.ImageWithCacheControl

Image with Anthropic prompt caching support.

methodinstructor.v2.core.multimodal.PDF.autodetect(source:str | Path) -> PDF

Attempt to autodetect a PDF from a source string or Path.

methodinstructor.v2.core.multimodal.PDF.from_gs_url(data_uri:str, timeout:int=30) -> PDF

Create a PDF instance from a Google Cloud Storage URL.

methodinstructor.v2.core.multimodal.PDF.to_bedrock(name:str | None=None) -> dict[str, Any]

Convert to Bedrock's document format.

classinstructor.v2.core.multimodal.PDFWithCacheControl

PDF with Anthropic prompt caching support.

func

instructor.v2.core.patch.apatch(client:AsyncOpenAI, mode:Mode=Mode.TOOLS, provider:Provider=Provider.OPENAI) -> AsyncOpenAI

Deprecated alias for :func:`patch`.

classinstructor.v2.core.protocols.RequestHandler

Prepares request kwargs for a specific mode.

classinstructor.v2.core.protocols.ResponseParser

Parses API response into validated Pydantic model.

classinstructor.v2.core.protocols.StreamExtractor

Extract JSON chunks from a streaming response.

classinstructor.v2.core.providers.Provider

Supported provider identifiers.

funcinstructor.v2.core.providers.get_provider(base_url:str) -> Provider

Detect the provider based on the base URL.

funcinstructor.v2.core.providers.provider_from_mode(mode:Mode, default:Provider=Provider.OPENAI) -> Provider

Infer provider from a provider-specific Mode.

classinstructor.v2.core.registry.ModeHandlers

Collection of handlers for a specific mode.

classinstructor.v2.core.registry.ModeRegistry

Central registry for mode handlers.

methodinstructor.v2.core.registry.ModeRegistry.get_handler_class(provider:Provider, mode:Mode) -> type | None

Get the handler class for a mode.

methodinstructor.v2.core.registry.ModeRegistry.get_handlers(provider:Provider, mode:Mode) -> ModeHandlers

Get all handlers for a mode.

methodinstructor.v2.core.registry.ModeRegistry.get_modes_for_provider(provider:Provider) -> list[Mode]

Get all registered modes for a provider.

methodinstructor.v2.core.registry.ModeRegistry.get_providers_for_mode(mode:Mode) -> list[Provider]

Get all providers that support a mode.

methodinstructor.v2.core.registry.ModeRegistry.is_registered(provider:Provider, mode:Mode) -> bool

Check if a mode is registered.

methodinstructor.v2.core.registry.ModeRegistry.list_modes() -> list[tuple[Provider, Mode]]

List all registered modes.

method

instructor.v2.core.registry.ModeRegistry.register_lazy(provider:Provider, mode:Mode, loader:Callable[[], ModeHandlers]) -> None

funcinstructor.v2.core.templating.apply_template(text:str, context:dict[str, Any]) -> str

Apply Jinja2 template to the given text.

classinstructor.v2.core.utils.classproperty

Descriptor for class-level properties.

funcinstructor.v2.core.utils.disable_pydantic_error_url() -> None

Disable URLs in Pydantic ValidationError messages.

funcinstructor.v2.core.utils.is_async(func:Callable[..., Any]) -> bool

Return whether a callable is async, following wrapped callables.

methodinstructor.v2.dsl.json_tracker.JsonCompleteness.get_complete_paths() -> set[str]

Return all paths that are complete.

methodinstructor.v2.dsl.json_tracker.JsonCompleteness.is_path_complete(path:str) -> bool

Check if the sub-structure at the given path is complete.

methodinstructor.v2.dsl.json_tracker.JsonCompleteness.is_root_complete() -> bool

Check if the root JSON structure is complete.

funcinstructor.v2.dsl.json_tracker.is_json_complete(json_str:str) -> bool

Check if a JSON string represents a complete structure.

funcinstructor.v2.dsl.maybe.Maybe(model:type[T]) -> type[MaybeBase[T]]

Create a Maybe model for a given Pydantic model.

classinstructor.v2.dsl.partial.PartialLiteralMixin

DEPRECATED: This mixin is no longer necessary.

classinstructor.v2.providers.anthropic.handlers.AnthropicHandlerBase

Common utilities for Anthropic handlers.

classinstructor.v2.providers.anthropic.handlers.AnthropicJSONHandler

Handler for Anthropic JSON mode.

classinstructor.v2.providers.anthropic.handlers.AnthropicParallelToolsHandler

Handler for Anthropic parallel tool calling.

classinstructor.v2.providers.anthropic.handlers.AnthropicToolsHandler

Handler for Anthropic TOOLS mode.

funcinstructor.v2.providers.anthropic.handlers.extract_system_messages(messages:list[dict[str, Any]]) -> list[SystemMessage]

Extract system messages from a list of messages.

funcinstructor.v2.providers.anthropic.usage.initialize_usage() -> Any

Create an empty Anthropic usage accumulator.

classinstructor.v2.providers.bedrock.handlers.BedrockMDJSONHandler

Handler for Bedrock MD_JSON mode.

classinstructor.v2.providers.bedrock.handlers.BedrockToolsHandler

Handler for Bedrock TOOLS mode.

funcinstructor.v2.providers.bedrock.handlers.generate_bedrock_schema(response_model:type[Any]) -> dict[str, Any]

Generate Bedrock tool schema from a Pydantic model.

func

instructor.v2.providers.bedrock.handlers.handle_bedrock_json(response_model:type[Any], new_kwargs:dict[str, Any]) -> tuple[type[Any], dict[str, Any]]

Handle Bedrock JSON mode.

func

instructor.v2.providers.bedrock.handlers.handle_bedrock_tools(response_model:type[Any] | None, new_kwargs:dict[str, Any]) -> tuple[type[Any] | None, dict[str, Any]]

Handle Bedrock tools mode.

classinstructor.v2.providers.cohere.handlers.CohereJSONSchemaHandler

Handler for Cohere JSON_SCHEMA mode.

classinstructor.v2.providers.cohere.handlers.CohereMDJSONHandler

Handler for Cohere MD_JSON mode.

classinstructor.v2.providers.cohere.handlers.CohereToolsHandler

Handler for Cohere TOOLS mode.

classinstructor.v2.providers.gemini.handlers.GeminiHandlerBase

Base handler for Gemini modes.

classinstructor.v2.providers.gemini.handlers.GeminiJSONHandler

Handler for Gemini JSON mode.

classinstructor.v2.providers.gemini.handlers.GeminiToolsHandler

Handler for Gemini TOOLS mode.

func

instructor.v2.providers.gemini.handlers.parse_gemini_json(response_model:type[BaseModel], completion:Any, validation_context:dict[str, Any] | None=None, strict:bool | None=None) -> BaseModel

Parse Gemini text JSON responses.

func

instructor.v2.providers.gemini.handlers.parse_gemini_tools(response_model:type[BaseModel], completion:Any, validation_context:dict[str, Any] | None=None, strict:bool | None=None) -> BaseModel

Parse Gemini tool-call responses.

funcinstructor.v2.providers.gemini.utils.reask_genai_tools(*args:Any, **kwargs:Any) -> dict[str, Any]

Compatibility shim for the GenAI-owned reask helper.

classinstructor.v2.providers.genai.handlers.GenAIHandlerBase

Common utilities shared across GenAI mode handlers.

classinstructor.v2.providers.genai.handlers.GenAIToolsHandler

Mode handler for GenAI tools/function calling.

func

instructor.v2.providers.genai.templating.process_message(message:Any, context:dict[str, Any], apply_template:Callable[[str, dict[str, Any]], str]) -> Any

Apply templates to GenAI Content parts.

funcinstructor.v2.providers.mistral.client.async_wrapper(*args:Any, **wrapper_kwargs:Any) -> Any

Async wrapper that handles streaming.

funcinstructor.v2.providers.mistral.client.sync_wrapper(*args:Any, **wrapper_kwargs:Any) -> Any

Sync wrapper that handles streaming.

classinstructor.v2.providers.mistral.handlers.MistralToolsHandler

Handler for Mistral TOOLS mode.

method

instructor.v2.providers.mistral.handlers.MistralToolsHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for tools mode.

classinstructor.v2.providers.openai.handlers.OpenAIParallelToolsHandler

Handler for OpenAI parallel tool calling.

classinstructor.v2.providers.openai.handlers.OpenAIResponsesToolsHandler

Handler for OpenAI Responses API with tools.

method

instructor.v2.providers.openai.handlers.OpenAIResponsesToolsHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for Responses API.

classinstructor.v2.providers.openai.handlers.OpenAIToolsHandler

Handler for OpenAI TOOLS mode.

method

instructor.v2.providers.openai.handlers.OpenAIToolsHandler.handle_reask(kwargs:dict[str, Any], response:ChatCompletion, exception:Exception) -> dict[str, Any]

Handle reask for tools mode.

method

instructor.v2.providers.openai.handlers.OpenAIToolsHandler.prepare_request(response_model:Any, kwargs:dict[str, Any]) -> tuple[Any, dict[str, Any]]

Prepare request with tool definitions.

classinstructor.v2.providers.openrouter.handlers.OpenRouterJSONSchemaHandler

Handler for OpenRouter structured outputs.

classinstructor.v2.providers.perplexity.handlers.PerplexityMDJSONHandler

Handler for Perplexity JSON mode.

func

instructor.v2.providers.perplexity.handlers.handle_perplexity_json(response_model:type[Any], new_kwargs:dict[str, Any]) -> tuple[type[Any], dict[str, Any]]

Handle Perplexity JSON mode.

classinstructor.v2.providers.vertexai.handlers.VertexAIHandlerBase

Base handler for VertexAI modes.

classinstructor.v2.providers.vertexai.handlers.VertexAIJSONHandler

Handler for VertexAI JSON mode.

classinstructor.v2.providers.vertexai.handlers.VertexAIParallelToolsHandler

Handler for VertexAI parallel tools mode.

classinstructor.v2.providers.vertexai.handlers.VertexAIToolsHandler

Handler for VertexAI TOOLS mode.

classinstructor.v2.providers.writer.handlers.WriterJSONSchemaHandler

Handler for Writer native JSON schema mode.

classinstructor.v2.providers.writer.handlers.WriterMDJSONHandler

Handler for Writer MD_JSON mode.

method

instructor.v2.providers.writer.handlers.WriterMDJSONHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for Writer MD_JSON mode.

classinstructor.v2.providers.writer.handlers.WriterToolsHandler

Handler for Writer TOOLS mode.

method

instructor.v2.providers.writer.handlers.WriterToolsHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for Writer tools mode.

func

instructor.v2.providers.writer.handlers.handle_writer_json(response_model:type[Any], new_kwargs:dict[str, Any]) -> tuple[type[Any], dict[str, Any]]

Handle Writer JSON mode.

func

instructor.v2.providers.writer.handlers.handle_writer_tools(response_model:type[Any], new_kwargs:dict[str, Any]) -> tuple[type[Any], dict[str, Any]]

Handle Writer tools mode.

classinstructor.v2.providers.xai.handlers.XAIHandlerBase

Base class for xAI handlers with shared utilities.

classinstructor.v2.providers.xai.handlers.XAIJSONSchemaHandler

Handler for xAI JSON_SCHEMA mode.

method

instructor.v2.providers.xai.handlers.XAIJSONSchemaHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for JSON schema mode.

classinstructor.v2.providers.xai.handlers.XAIMDJSONHandler

Handler for xAI MD_JSON mode.

method

instructor.v2.providers.xai.handlers.XAIMDJSONHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for MD_JSON mode.

classinstructor.v2.providers.xai.handlers.XAIParallelToolsHandler

Handler for xAI parallel tool calling.

method

instructor.v2.providers.xai.handlers.XAIParallelToolsHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for parallel tools mode.

classinstructor.v2.providers.xai.handlers.XAIToolsHandler

Handler for xAI TOOLS mode.

method

instructor.v2.providers.xai.handlers.XAIToolsHandler.handle_reask(kwargs:dict[str, Any], response:Any, exception:Exception) -> dict[str, Any]

Handle reask for tools mode.

func

instructor.v2.providers.xai.handlers.handle_xai_json(response_model:type[Any] | None, new_kwargs:dict[str, Any]) -> tuple[type[Any] | None, dict[str, Any]]

Handle xAI JSON mode.

func

instructor.v2.providers.xai.handlers.handle_xai_tools(response_model:type[Any] | None, new_kwargs:dict[str, Any]) -> tuple[type[Any] | None, dict[str, Any]]

Handle xAI tools mode.

classinstructor.v2.validation.async_validators.AsyncValidationContext

Carry context through async validation hooks.

funcinstructor.v2.validation.async_validators.async_field_validator(field:str, *fields:str) -> Callable[[T], T]

Mark a callable as an async field validator.

About this data

These signatures were extracted from the public source of instructor-ai/instructor using Python's ast module. Argument names, default values, type annotations and return types are taken verbatim from the code. Implementation bodies are never stored. See how it works for details.

Back to all 805 libraries