langgraph API reference

321 public APIs from langgraph (langchain-ai/langgraph) — 163 classes, 56 functions, 102 methods. Signatures extracted by static analysis of the actual source.

Repository: langchain-ai/langgraph

Kind	Count
Classes	163
Functions	56
Methods	102

API list

classlibs.checkpoint-conformance.langgraph.checkpoint.conformance.report.CapabilityReport

Aggregate report across all capabilities.

methodlibs.checkpoint-conformance.langgraph.checkpoint.conformance.report.CapabilityReport.print_report() -> None

Pretty-print the report to stdout.

methodlibs.checkpoint-conformance.langgraph.checkpoint.conformance.report.CapabilityReport.to_dict() -> dict[str, Any]

Return a JSON-serializable dict.

classlibs.checkpoint-sqlite.langgraph.cache.sqlite.SqliteCache

File-based cache using SQLite.

methodlibs.checkpoint-sqlite.langgraph.cache.sqlite.SqliteCache.clear(namespaces:Sequence[Namespace] | None=None) -> None

Delete the cached values for the given namespaces.

methodlibs.checkpoint-sqlite.langgraph.cache.sqlite.SqliteCache.get(keys:Sequence[FullKey]) -> dict[FullKey, ValueT]

Get the cached values for the given keys.

method

libs.checkpoint-sqlite.langgraph.cache.sqlite.SqliteCache.set(mapping:Mapping[FullKey, tuple[ValueT, int | None]]) -> None

Set the cached values for the given keys and TTLs.

classlibs.checkpoint-sqlite.langgraph.store.sqlite.base.BaseSqliteStore

Shared base class for SQLite stores.

classlibs.checkpoint.langgraph.cache.base.BaseCache

Base class for a cache.

methodlibs.checkpoint.langgraph.cache.base.BaseCache.clear(namespaces:Sequence[Namespace] | None=None) -> None

Delete the cached values for the given namespaces.

methodlibs.checkpoint.langgraph.cache.base.BaseCache.get(keys:Sequence[FullKey]) -> dict[FullKey, ValueT]

Get the cached values for the given keys.

methodlibs.checkpoint.langgraph.cache.base.BaseCache.set(pairs:Mapping[FullKey, tuple[ValueT, int | None]]) -> None

Set the cached values for the given keys and TTLs.

methodlibs.checkpoint.langgraph.cache.memory.InMemoryCache.clear(namespaces:Sequence[Namespace] | None=None) -> None

Delete the cached values for the given namespaces.

methodlibs.checkpoint.langgraph.cache.memory.InMemoryCache.get(keys:Sequence[FullKey]) -> dict[FullKey, ValueT]

Get the cached values for the given keys.

methodlibs.checkpoint.langgraph.cache.memory.InMemoryCache.set(keys:Mapping[FullKey, tuple[ValueT, int | None]]) -> None

Set the cached values for the given keys.

classlibs.checkpoint.langgraph.cache.redis.RedisCache

Redis-based cache implementation with TTL support.

methodlibs.checkpoint.langgraph.cache.redis.RedisCache.clear(namespaces:Sequence[Namespace] | None=None) -> None

Delete the cached values for the given namespaces.

methodlibs.checkpoint.langgraph.cache.redis.RedisCache.get(keys:Sequence[FullKey]) -> dict[FullKey, ValueT]

Get the cached values for the given keys.

methodlibs.checkpoint.langgraph.cache.redis.RedisCache.set(mapping:Mapping[FullKey, tuple[ValueT, int | None]]) -> None

Set the cached values for the given keys and TTLs.

classlibs.checkpoint.langgraph.checkpoint.base.BaseCheckpointSaver

Base class for creating a graph checkpointer.

methodlibs.checkpoint.langgraph.checkpoint.base.BaseCheckpointSaver.get(config:RunnableConfig) -> Checkpoint | None

Fetch a checkpoint using the given configuration.

methodlibs.checkpoint.langgraph.checkpoint.base.BaseCheckpointSaver.get_next_version(current:V | None, channel:None) -> V

Generate the next version ID for a channel.

methodlibs.checkpoint.langgraph.checkpoint.base.BaseCheckpointSaver.get_tuple(config:RunnableConfig) -> CheckpointTuple | None

Fetch a checkpoint tuple using the given configuration.

method

libs.checkpoint.langgraph.checkpoint.base.BaseCheckpointSaver.prune(thread_ids:Sequence[str], *strategy:str='keep_latest') -> None

Prune checkpoints for the given threads.

classlibs.checkpoint.langgraph.checkpoint.base.Checkpoint

State snapshot at a given point in time.

classlibs.checkpoint.langgraph.checkpoint.base.CheckpointMetadata

Metadata associated with a checkpoint.

funclibs.checkpoint.langgraph.checkpoint.base.get_checkpoint_id(config:RunnableConfig) -> str | None

Get checkpoint ID.

classlibs.checkpoint.langgraph.checkpoint.base.id.UUID

UUID draft version objects

classlibs.checkpoint.langgraph.checkpoint.memory.InMemorySaver

An in-memory checkpoint saver.

methodlibs.checkpoint.langgraph.checkpoint.memory.InMemorySaver.aget_tuple(config:RunnableConfig) -> CheckpointTuple | None

Asynchronous version of `get_tuple`.

method

libs.checkpoint.langgraph.checkpoint.memory.InMemorySaver.aput(config:RunnableConfig, checkpoint:Checkpoint, metadata:CheckpointMetadata, new_versions:ChannelVersions) -> RunnableConfig

Asynchronous version of `put`.

method

libs.checkpoint.langgraph.checkpoint.memory.InMemorySaver.aput_writes(config:RunnableConfig, writes:Sequence[tuple[str, Any]], task_id:str, task_path:str='') -> None

Asynchronous version of `put_writes`.

methodlibs.checkpoint.langgraph.checkpoint.memory.InMemorySaver.get_tuple(config:RunnableConfig) -> CheckpointTuple | None

Get a checkpoint tuple from the in-memory storage.

classlibs.checkpoint.langgraph.checkpoint.serde.base.CipherProtocol

Protocol for encryption and decryption of data.

methodlibs.checkpoint.langgraph.checkpoint.serde.base.CipherProtocol.decrypt(ciphername:str, ciphertext:bytes) -> bytes

Decrypt ciphertext.

methodlibs.checkpoint.langgraph.checkpoint.serde.base.CipherProtocol.encrypt(plaintext:bytes) -> tuple[str, bytes]

Encrypt plaintext.

funclibs.checkpoint.langgraph.checkpoint.serde.event_hooks.emit_serde_event(event:SerdeEvent) -> None

Emit a serde event to all listeners.

classlibs.checkpoint.langgraph.store.base.BaseStore

Abstract base class for persistent key-value stores.

methodlibs.checkpoint.langgraph.store.base.BaseStore.abatch(ops:Iterable[Op]) -> list[Result]

Execute multiple operations asynchronously in a single batch.

methodlibs.checkpoint.langgraph.store.base.BaseStore.adelete(namespace:tuple[str, ...], key:str) -> None

Asynchronously delete an item.

method

libs.checkpoint.langgraph.store.base.BaseStore.aget(namespace:tuple[str, ...], key:str, *refresh_ttl:bool | None=None) -> Item | None

Asynchronously retrieve a single item.

methodlibs.checkpoint.langgraph.store.base.BaseStore.batch(ops:Iterable[Op]) -> list[Result]

Execute multiple operations synchronously in a single batch.

methodlibs.checkpoint.langgraph.store.base.BaseStore.delete(namespace:tuple[str, ...], key:str) -> None

Delete an item.

method

libs.checkpoint.langgraph.store.base.BaseStore.get(namespace:tuple[str, ...], key:str, *refresh_ttl:bool | None=None) -> Item | None

Retrieve a single item.

classlibs.checkpoint.langgraph.store.base.InvalidNamespaceError

Provided namespace is invalid.

classlibs.checkpoint.langgraph.store.base.Item

Represents a stored item with metadata.

classlibs.checkpoint.langgraph.store.base.ListNamespacesOp

Operation to list and filter namespaces in the store.

classlibs.checkpoint.langgraph.store.base.NotProvided

Sentinel singleton.

classlibs.checkpoint.langgraph.store.base.PutOp

Operation to store, update, or delete an item in the store.

funclibs.checkpoint.langgraph.store.base.embed.tokenize_path(path:str) -> list[str]

Tokenize a path into components.

funclibs.cli.langgraph_cli.cli.new(path:str | None, template:str | None) -> None

Create a new LangGraph project from a template.

funclibs.cli.langgraph_cli.config.validate_config(config:Config) -> Config

Validate a configuration dictionary.

funclibs.cli.langgraph_cli.config.validate_config_file(config_path:pathlib.Path) -> Config

Load and validate a configuration file.

funclibs.cli.langgraph_cli.deploy.format_log_entry(entry:dict) -> str

Format a single log entry for display.

funclibs.cli.langgraph_cli.deploy.level_fg(level:str) -> str | None

Return click color for a log level.

funclibs.cli.langgraph_cli.deploy.normalize_image_tag(value:str) -> str

Validate and return a Docker image tag.

funclibs.cli.langgraph_cli.deploy.validate_deployment_selector(deployment_id:str | None, name:str | None) -> None

Ensure either deployment_id or name is provided.

funclibs.cli.langgraph_cli.docker.dict_to_yaml(d:dict, *indent:int=0) -> str

Convert a dictionary to a YAML string.

classlibs.cli.langgraph_cli.schemas.EncryptionConfig

Configuration for custom at-rest encryption logic.

classlibs.cli.langgraph_cli.schemas.GraphDef

Definition of a graph with additional metadata.

classlibs.cli.langgraph_cli.schemas.StoreConfig

Configuration for the built-in long-term memory store.

classlibs.cli.langgraph_cli.schemas.UvSource

Deployment source rooted at a uv project or workspace.

funclibs.cli.langgraph_cli.util.warn_non_wolfi_distro(config_json:dict, *emit:Callable[[str], None] | None=None) -> None

Show warning if image_distro is not set to 'wolfi'.

funclibs.cli.python-monorepo-example.apps.agent.src.agent.graph.call_model(state:State) -> dict

Simple node that uses the shared libraries.

classlibs.cli.python-monorepo-example.apps.agent.src.agent.state.State

The state of the agent.

funclibs.cli.python-monorepo-example.libs.common.helpers.get_common_prefix() -> str

Get a common prefix for messages.

funclibs.cli.python-monorepo-example.libs.shared.src.shared.utils.get_dummy_message() -> str

Get a dummy message for testing.

funclibs.langgraph.bench.__main__.arun_first_event_latency(graph:Pregel, input:dict) -> None

Latency for the first event.

funclibs.langgraph.bench.__main__.compile_graph(graph:StateGraph) -> None

Compile the graph.

funclibs.langgraph.bench.__main__.run_first_event_latency(graph:Pregel, input:dict) -> None

Latency for the first event.

funclibs.langgraph.bench.sequential.anoop(state:MessagesState) -> None

No-op function.

funclibs.langgraph.bench.sequential.noop(state:MessagesState) -> None

No-op function.

func

libs.langgraph.langgraph._internal._config.get_callback_manager_for_config(config:RunnableConfig, tags:Sequence[str] | None=None) -> CallbackManager

Get a callback manager for a config.

funclibs.langgraph.langgraph._internal._config.merge_configs(*configs:RunnableConfig | None) -> RunnableConfig

Merge multiple configs into one.

funclibs.langgraph.langgraph._internal._config.recast_checkpoint_ns(ns:str) -> str

Remove task IDs from checkpoint namespace.

funclibs.langgraph.langgraph._internal._fields.get_cached_annotated_keys(obj:type[Any]) -> tuple[str, ...]

Return cached annotated keys for a Python class.

classlibs.langgraph.langgraph._internal._queue.AsyncQueue

Async unbounded FIFO queue with a wait() method.

methodlibs.langgraph.langgraph._internal._queue.AsyncQueue.wait() -> None

If queue is empty, wait until an item is available.

classlibs.langgraph.langgraph._internal._queue.Semaphore

Semaphore subclass with a wait() method.

classlibs.langgraph.langgraph._internal._queue.SyncQueue

Unbounded FIFO queue with a wait() method.

classlibs.langgraph.langgraph._internal._runnable.StrEnum

A string enum.

func

libs.langgraph.langgraph._internal._runnable.coerce_to_runnable(thing:RunnableLike, *name:str | None, *trace:bool) -> Runnable

Coerce a runnable-like object into a Runnable.

funclibs.langgraph.langgraph._internal._runnable.is_async_callable(func:Any) -> TypeGuard[Callable[..., Awaitable]]

Check if a function is async.

funclibs.langgraph.langgraph._internal._runnable.is_async_generator(func:Any) -> TypeGuard[Callable[..., AsyncIterator]]

Check if a function is an async generator.

classlibs.langgraph.langgraph.callbacks.GraphCallbackHandler

Base class for graph-level lifecycle callbacks.

methodlibs.langgraph.langgraph.channels.any_value.AnyValue.ValueType() -> type[Value]

The type of the value stored in the channel.

methodlibs.langgraph.langgraph.channels.any_value.AnyValue.copy() -> Self

Return a copy of the channel.

classlibs.langgraph.langgraph.channels.base.BaseChannel

Base class for all channels.

methodlibs.langgraph.langgraph.channels.base.BaseChannel.UpdateType() -> Any

The type of the update received by the channel.

methodlibs.langgraph.langgraph.channels.base.BaseChannel.ValueType() -> Any

The type of the value stored in the channel.

methodlibs.langgraph.langgraph.channels.base.BaseChannel.consume() -> bool

Notify the channel that a subscribed task ran.

methodlibs.langgraph.langgraph.channels.base.BaseChannel.copy() -> Self

Return a copy of the channel.

methodlibs.langgraph.langgraph.channels.base.BaseChannel.get() -> Value

Return the current value of the channel.

classlibs.langgraph.langgraph.channels.topic.Topic

A configurable PubSub Topic.

methodlibs.langgraph.langgraph.channels.topic.Topic.UpdateType() -> Any

The type of the update received by the channel.

methodlibs.langgraph.langgraph.channels.topic.Topic.ValueType() -> Any

The type of the value stored in the channel.

methodlibs.langgraph.langgraph.channels.topic.Topic.copy() -> Self

Return a copy of the channel.

classlibs.langgraph.langgraph.errors.EmptyInputError

Raised when graph receives an empty input.

classlibs.langgraph.langgraph.errors.NodeError

Failure context passed to a node-level error handler.

classlibs.langgraph.langgraph.errors.NodeInterrupt

Raised by a node to interrupt execution.

classlibs.langgraph.langgraph.func.entrypoint.final

A primitive that can be returned from an entrypoint.

classlibs.langgraph.langgraph.graph.ui.UIMessage

A message type for UI updates in LangGraph.

funclibs.langgraph.langgraph.graph.ui.delete_ui_message(id:str, *state_key:str='ui') -> RemoveUIMessage

Delete a UI message by ID from the UI state.

funclibs.langgraph.langgraph.pregel._algo.checkpoint_null_version(checkpoint:Checkpoint) -> V | None

Get the null version for the checkpoint, if available.

funclibs.langgraph.langgraph.pregel._algo.task_path_str(tup:str | int | tuple) -> str

Generate a string representation of the task path.

funclibs.langgraph.langgraph.pregel._call.identifier(obj:Any, name:str | None=None) -> str | None

Return the module and name of an object.

funclibs.langgraph.langgraph.pregel._checkpoint.exit_delta_task_id(step:int, task_id:str) -> str

Synthetic task id for exit-mode DeltaChannel writes.

func

libs.langgraph.langgraph.pregel._draw.add_edge(graph:Graph, source:str, target:str, *data:Any | None=None, *conditional:bool=False) -> None

Add an edge to the graph.

classlibs.langgraph.langgraph.pregel._messages.StreamMessagesHandlerV2

v2 variant of `StreamMessagesHandler`.

classlibs.langgraph.langgraph.pregel._read.PregelNode

A node in a Pregel graph.

methodlibs.langgraph.langgraph.pregel._read.PregelNode.flat_writers() -> list[Runnable]

Get writers with optimizations applied.

methodlibs.langgraph.langgraph.pregel._read.PregelNode.input_cache_key() -> INPUT_CACHE_KEY_TYPE

Get a cache key for the input to the node.

func

libs.langgraph.langgraph.pregel._retry.run_with_retry(task:PregelExecutableTask, retry_policy:Sequence[RetryPolicy] | None, configurable:dict[str, Any] | None=None) -> None

Run a task with retries.

classlibs.langgraph.langgraph.pregel._utils.FunctionNonLocals

Get the nonlocal variables accessed of a function.

methodlibs.langgraph.langgraph.pregel._utils.FunctionNonLocals.visit_AsyncFunctionDef(node:ast.AsyncFunctionDef) -> Any

Visit an async function definition.

methodlibs.langgraph.langgraph.pregel._utils.FunctionNonLocals.visit_FunctionDef(node:ast.FunctionDef) -> Any

Visit a function definition.

methodlibs.langgraph.langgraph.pregel._utils.FunctionNonLocals.visit_Lambda(node:ast.Lambda) -> Any

Visit a lambda function.

classlibs.langgraph.langgraph.pregel._utils.NonLocals

Get nonlocal variables accessed.

methodlibs.langgraph.langgraph.pregel._utils.NonLocals.visit_Attribute(node:ast.Attribute) -> Any

Visit an attribute node.

methodlibs.langgraph.langgraph.pregel._utils.NonLocals.visit_Name(node:ast.Name) -> Any

Visit a name node.

funclibs.langgraph.langgraph.pregel._utils.get_function_nonlocals(func:Callable) -> list[Any]

Get the nonlocal variables accessed by a function.

funclibs.langgraph.langgraph.pregel.debug.get_bolded_text(text:str) -> str

Get bolded text.

funclibs.langgraph.langgraph.pregel.debug.get_colored_text(text:str, color:str) -> str

Get colored text.

funclibs.langgraph.langgraph.pregel.debug.map_debug_tasks(tasks:Iterable[PregelExecutableTask]) -> Iterator[TaskPayload]

Produce "task" events for stream_mode=debug.

funclibs.langgraph.langgraph.pregel.debug.rm_pregel_keys(config:RunnableConfig | None) -> RunnableConfig | None

Remove pregel-specific keys from the config.

classlibs.langgraph.langgraph.runtime.RunControl

Run-scoped control surface for cooperative draining.

methodlibs.langgraph.langgraph.runtime.Runtime.merge(other:Runtime[ContextT]) -> Runtime[ContextT]

Merge two runtimes together.

classlibs.langgraph.langgraph.runtime.ServerInfo

Metadata injected by LangGraph Server.

funclibs.langgraph.langgraph.runtime.get_runtime(context_schema:type[ContextT] | None=None) -> Runtime[ContextT]

Get the runtime for the current graph run.

funclibs.langgraph.langgraph.stream._convert.convert_to_protocol_event(part:StreamPart) -> ProtocolEvent

Convert a v2 StreamPart to a ProtocolEvent.

classlibs.langgraph.langgraph.stream._types.StreamTransformer

Extension point for custom stream projections.

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.afail(err:BaseException) -> None

Called when the run ends with an error (async lane).

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.afinalize() -> None

Called when the run ends normally (async lane).

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.aprocess(event:ProtocolEvent) -> bool

Handle an event on the async lane.

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.fail(err:BaseException) -> None

Called when the run ends with an error (sync lane).

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.finalize() -> None

Called when the run ends normally (sync lane).

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.init() -> dict[str, Any]

Return the projection dict.

methodlibs.langgraph.langgraph.stream._types.StreamTransformer.process(event:ProtocolEvent) -> bool

Handle an event on the sync lane.

classlibs.langgraph.langgraph.stream.run_stream.AsyncGraphRunStream

Async run stream with caller-driven pumping.

methodlibs.langgraph.langgraph.stream.run_stream.AsyncGraphRunStream.abort() -> None

Stop the run early.

classlibs.langgraph.langgraph.stream.run_stream.GraphRunStream

Sync run stream with caller-driven pumping.

methodlibs.langgraph.langgraph.stream.run_stream.GraphRunStream.abort() -> None

Stop the run early.

classlibs.langgraph.langgraph.stream.transformers.DebugTransformer

Capture debug events as a drainable stream.

classlibs.langgraph.langgraph.stream.transformers.TasksTransformer

Capture raw task events as a drainable stream.

classlibs.langgraph.langgraph.types.CacheKey

Cache key for a task.

classlibs.langgraph.langgraph.types.CachePolicy

Configuration for caching nodes.

classlibs.langgraph.langgraph.types.CheckpointPayload

Payload for a checkpoint event.

classlibs.langgraph.langgraph.types.CheckpointStreamPart

Stream part emitted for `stream_mode="checkpoints"`.

classlibs.langgraph.langgraph.types.CheckpointTask

A task entry within a `CheckpointPayload`.

classlibs.langgraph.langgraph.types.CustomStreamPart

Stream part emitted for `stream_mode="custom"`.

classlibs.langgraph.langgraph.types.DebugStreamPart

Stream part emitted for `stream_mode="debug"`.

classlibs.langgraph.langgraph.types.Interrupt

Information about an interrupt that occurred in a node.

classlibs.langgraph.langgraph.types.MessagesStreamPart

Stream part emitted for `stream_mode="messages"`.

classlibs.langgraph.langgraph.types.PregelTask

A Pregel task.

classlibs.langgraph.langgraph.types.RetryPolicy

Configuration for retrying nodes.

classlibs.langgraph.langgraph.types.TaskPayload

Payload for a task start event.

classlibs.langgraph.langgraph.types.TaskResultPayload

Payload for a task result event.

classlibs.langgraph.langgraph.types.TasksStreamPart

Stream part emitted for `stream_mode="tasks"`.

classlibs.langgraph.langgraph.types.TimeoutPolicy

Configuration for timing out node attempts.

classlibs.langgraph.langgraph.types.UpdatesStreamPart

Stream part emitted for `stream_mode="updates"`.

classlibs.langgraph.langgraph.types.ValuesStreamPart

Stream part emitted for `stream_mode="values"`.

funclibs.langgraph.langgraph.types.interrupt(value:Any) -> Any

Interrupt the graph with a resumable exception from within a node.

classlibs.langgraph.langgraph.warnings.LangGraphDeprecationWarning

A LangGraph specific deprecation warning.

classlibs.prebuilt.langgraph.prebuilt._tool_call_stream.ToolCallStream

Scoped view of a single tool call's lifecycle.

classlibs.prebuilt.langgraph.prebuilt.chat_agent_executor.AgentState

The state of the agent.

classlibs.prebuilt.langgraph.prebuilt.chat_agent_executor.AgentStatePydantic

The state of the agent.

funclibs.prebuilt.langgraph.prebuilt.chat_agent_executor.post_model_hook_router(state:StateSchema) -> str | list[Send]

Route to the next node after post_model_hook.

classlibs.prebuilt.langgraph.prebuilt.tool_node.ToolCallWithContext

ToolCall with additional context for graph state.

classlibs.prebuilt.langgraph.prebuilt.tool_node.ToolNode

A node for executing tools in LangGraph workflows.

funclibs.prebuilt.langgraph.prebuilt.tool_node.ToolNode.execute(req:ToolCallRequest) -> ToolMessage | Command

Execute tool with given request.

classlibs.prebuilt.langgraph.prebuilt.tool_node.ToolRuntime

Runtime context automatically injected into tools.

funclibs.prebuilt.langgraph.prebuilt.tool_node.msg_content_output(output:Any) -> str | list[dict]

Convert tool output to `ToolMessage` content format.

classlibs.sdk-py.integration.graph.streaming_graph.AgentState

Top-level state for the agent.

funclibs.sdk-py.integration.graph.streaming_graph.ask_human(state:AgentState) -> dict[str, Any]

Pause the graph and wait for a `thread.run.respond(...)`.

funclibs.sdk-py.integration.graph.streaming_graph.call_tool(state:AgentState) -> dict[str, Any]

Invoke a tool and emit its result as a tool message.

funclibs.sdk-py.integration.graph.streaming_graph.search(query:str) -> str

Look up `query` in a fake search index.

funclibs.sdk-py.integration.graph.streaming_graph.stream_message(state:AgentState) -> dict[str, Any]

Stream an AI message via a fake chat model.

funclibs.sdk-py.integration.graph.tools_agent.search(query:str) -> str

Look up `query` in a fake search index.

classlibs.sdk-py.langgraph_sdk._async.assistants.AssistantsClient

Client for managing assistants in LangGraph.

method

libs.sdk-py.langgraph_sdk._async.assistants.AssistantsClient.delete(assistant_id:str, *delete_threads:bool=False, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Delete an assistant.

method

libs.sdk-py.langgraph_sdk._async.assistants.AssistantsClient.get(assistant_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> Assistant

Get an assistant by ID.

classlibs.sdk-py.langgraph_sdk._async.client.LangGraphClient

Top-level client for LangGraph API.

methodlibs.sdk-py.langgraph_sdk._async.client.LangGraphClient.aclose() -> None

Close the underlying HTTP client.

classlibs.sdk-py.langgraph_sdk._async.http.HttpClient

Handle async requests to the LangGraph API.

method

libs.sdk-py.langgraph_sdk._async.http.HttpClient.delete(path:str, *json:Any | None=None, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> None

Send a `DELETE` request.

method

libs.sdk-py.langgraph_sdk._async.http.HttpClient.get(path:str, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `GET` request.

method

libs.sdk-py.langgraph_sdk._async.http.HttpClient.patch(path:str, *json:dict, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `PATCH` request.

method

libs.sdk-py.langgraph_sdk._async.http.HttpClient.post(path:str, *json:dict[str, Any] | list | None, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `POST` request.

method

libs.sdk-py.langgraph_sdk._async.http.HttpClient.put(path:str, *json:dict, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `PUT` request.

classlibs.sdk-py.langgraph_sdk._async.runs.RunsClient

Client for managing runs in LangGraph.

method

libs.sdk-py.langgraph_sdk._async.runs.RunsClient.cancel(thread_id:str, run_id:str, *wait:bool=False, *action:CancelAction='interrupt', *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Get a run.

method

libs.sdk-py.langgraph_sdk._async.runs.RunsClient.delete(thread_id:str, run_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Delete a run.

method

libs.sdk-py.langgraph_sdk._async.runs.RunsClient.get(thread_id:str, run_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> Run

Get a run.

method

libs.sdk-py.langgraph_sdk._async.runs.RunsClient.join(thread_id:str, run_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> dict

Block until a run is done.

method

libs.sdk-py.langgraph_sdk._async.runs.RunsClient.list(thread_id:str, *limit:int=10, *offset:int=0, *status:RunStatus | None=None, *select:builtins.list[RunSelectField] | None=None, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> builtins.list[Run]

List runs.

funclibs.sdk-py.langgraph_sdk._async.runs.RunsClient.on_response(res:httpx.Response)

Callback function to handle the response.

method

libs.sdk-py.langgraph_sdk._async.store.StoreClient.delete_item(namespace:Sequence[str], key:str, headers:Mapping[str, str] | None=None, params:QueryParamTypes | None=None) -> None

Delete an item.

method

libs.sdk-py.langgraph_sdk._async.store.StoreClient.get_item(namespace:Sequence[str], key:str, *refresh_ttl:bool | None=None, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> Item

Retrieve a single item.

classlibs.sdk-py.langgraph_sdk._async.stream.RunModule

Command dispatcher for `run.start`.

method

libs.sdk-py.langgraph_sdk._async.stream.RunModule.start(*input:Any=None, *config:dict[str, Any] | None=None, *metadata:dict[str, Any] | None=None) -> dict[str, Any]

Send `run.start` to the server.

classlibs.sdk-py.langgraph_sdk._async.stream.ToolCallHandle

Async handle for one root-scope tool call.

classlibs.sdk-py.langgraph_sdk._async.threads.ThreadsClient

Client for managing threads in LangGraph.

method

libs.sdk-py.langgraph_sdk._async.threads.ThreadsClient.copy(thread_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Copy a thread.

method

libs.sdk-py.langgraph_sdk._async.threads.ThreadsClient.delete(thread_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Delete a thread.

method

libs.sdk-py.langgraph_sdk._async.threads.ThreadsClient.get(thread_id:str, *include:Sequence[str] | None=None, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> Thread

Get a thread by ID.

method

libs.sdk-py.langgraph_sdk._async.threads.ThreadsClient.prune(thread_ids:Sequence[str], *strategy:PruneStrategy='delete', *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> dict[str, Any]

Prune threads by ID.

method

libs.sdk-py.langgraph_sdk._sync.cron.SyncCronClient.delete(cron_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Delete a cron.

classlibs.sdk-py.langgraph_sdk._sync.http.SyncHttpClient

Handle synchronous requests to the LangGraph API.

method

libs.sdk-py.langgraph_sdk._sync.http.SyncHttpClient.get(path:str, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `GET` request.

method

libs.sdk-py.langgraph_sdk._sync.http.SyncHttpClient.patch(path:str, *json:dict, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `PATCH` request.

method

libs.sdk-py.langgraph_sdk._sync.http.SyncHttpClient.post(path:str, *json:dict[str, Any] | list | None, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `POST` request.

method

libs.sdk-py.langgraph_sdk._sync.http.SyncHttpClient.put(path:str, *json:dict, *params:QueryParamTypes | None=None, *headers:Mapping[str, str] | None=None, *on_response:Callable[[httpx.Response], None] | None=None) -> Any

Send a `PUT` request.

classlibs.sdk-py.langgraph_sdk._sync.runs.SyncRunsClient

Synchronous client for managing runs in LangGraph.

method

libs.sdk-py.langgraph_sdk._sync.runs.SyncRunsClient.cancel(thread_id:str, run_id:str, *wait:bool=False, *action:CancelAction='interrupt', *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Get a run.

method

libs.sdk-py.langgraph_sdk._sync.runs.SyncRunsClient.delete(thread_id:str, run_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Delete a run.

method

libs.sdk-py.langgraph_sdk._sync.runs.SyncRunsClient.get(thread_id:str, run_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> Run

Get a run.

method

libs.sdk-py.langgraph_sdk._sync.runs.SyncRunsClient.join(thread_id:str, run_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> dict

Block until a run is done.

method

libs.sdk-py.langgraph_sdk._sync.runs.SyncRunsClient.list(thread_id:str, *limit:int=10, *offset:int=0, *status:RunStatus | None=None, *select:builtins.list[RunSelectField] | None=None, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> builtins.list[Run]

List runs.

classlibs.sdk-py.langgraph_sdk._sync.stream.SyncRunModule

Command dispatcher for `run.start`.

method

libs.sdk-py.langgraph_sdk._sync.stream.SyncRunModule.start(*input:Any=None, *config:dict[str, Any] | None=None, *metadata:dict[str, Any] | None=None) -> dict[str, Any]

Send `run.start` to the server.

classlibs.sdk-py.langgraph_sdk._sync.stream.SyncToolCallHandle

Sync handle for one root-scope tool call.

method

libs.sdk-py.langgraph_sdk._sync.threads.SyncThreadsClient.copy(thread_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Copy a thread.

method

libs.sdk-py.langgraph_sdk._sync.threads.SyncThreadsClient.delete(thread_id:str, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> None

Delete a thread.

method

libs.sdk-py.langgraph_sdk._sync.threads.SyncThreadsClient.get(thread_id:str, *include:Sequence[str] | None=None, *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> Thread

Get a thread by ID.

method

libs.sdk-py.langgraph_sdk._sync.threads.SyncThreadsClient.prune(thread_ids:Sequence[str], *strategy:PruneStrategy='delete', *headers:Mapping[str, str] | None=None, *params:QueryParamTypes | None=None) -> dict[str, Any]

Prune threads by ID.

methodlibs.sdk-py.langgraph_sdk.auth.Auth.authenticate(fn:AH) -> AH

classlibs.sdk-py.langgraph_sdk.auth.types.AssistantsCreate

Payload for creating an assistant.

classlibs.sdk-py.langgraph_sdk.auth.types.AssistantsDelete

Payload for deleting an assistant.

classlibs.sdk-py.langgraph_sdk.auth.types.AssistantsRead

Payload for reading an assistant.

classlibs.sdk-py.langgraph_sdk.auth.types.AssistantsSearch

Payload for searching assistants.

classlibs.sdk-py.langgraph_sdk.auth.types.AssistantsUpdate

Payload for updating an assistant.

classlibs.sdk-py.langgraph_sdk.auth.types.BaseAuthContext

Base class for authentication context.

classlibs.sdk-py.langgraph_sdk.auth.types.BaseUser

The base ASGI user protocol

methodlibs.sdk-py.langgraph_sdk.auth.types.BaseUser.display_name() -> str

The display name of the user.

methodlibs.sdk-py.langgraph_sdk.auth.types.BaseUser.identity() -> str

The unique identifier for the user.

methodlibs.sdk-py.langgraph_sdk.auth.types.BaseUser.is_authenticated() -> bool

Whether the user is authenticated.

methodlibs.sdk-py.langgraph_sdk.auth.types.BaseUser.permissions() -> Sequence[str]

The permissions associated with the user.

classlibs.sdk-py.langgraph_sdk.auth.types.CronsCreate

Payload for creating a cron job.

classlibs.sdk-py.langgraph_sdk.auth.types.CronsDelete

Payload for deleting a cron job.

classlibs.sdk-py.langgraph_sdk.auth.types.CronsRead

Payload for reading a cron job.

classlibs.sdk-py.langgraph_sdk.auth.types.CronsSearch

Payload for searching cron jobs.

classlibs.sdk-py.langgraph_sdk.auth.types.CronsUpdate

Payload for updating a cron job.

methodlibs.sdk-py.langgraph_sdk.auth.types.MinimalUser.identity() -> str

The unique identifier for the user.

classlibs.sdk-py.langgraph_sdk.auth.types.MinimalUserDict

The dictionary representation of a user.

classlibs.sdk-py.langgraph_sdk.auth.types.RunsCreate

Payload for creating a run.

classlibs.sdk-py.langgraph_sdk.auth.types.StoreDelete

Operation to delete an item from the store.

classlibs.sdk-py.langgraph_sdk.auth.types.ThreadTTL

Time-to-live configuration for a thread.

classlibs.sdk-py.langgraph_sdk.auth.types.ThreadsCreate

Parameters for creating a new thread.

classlibs.sdk-py.langgraph_sdk.auth.types.ThreadsDelete

Parameters for deleting a thread.

classlibs.sdk-py.langgraph_sdk.auth.types.ThreadsSearch

Parameters for searching threads.

classlibs.sdk-py.langgraph_sdk.auth.types.ThreadsUpdate

Parameters for updating a thread or run.

classlibs.sdk-py.langgraph_sdk.auth.types.on.assistants

Types for assistant-related operations.

classlibs.sdk-py.langgraph_sdk.auth.types.on.assistants.create

Type for assistant creation parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.assistants.delete

Type for assistant deletion parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.assistants.read

Type for assistant read parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.assistants.search

Type for assistant search parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.assistants.update

Type for assistant update parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.crons

Types for cron-related operations.

classlibs.sdk-py.langgraph_sdk.auth.types.on.crons.create

Type for cron creation parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.crons.delete

Type for cron deletion parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.crons.read

Type for cron read parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.crons.search

Type for cron search parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.crons.update

Type for cron update parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.store

Types for store-related operations.

classlibs.sdk-py.langgraph_sdk.auth.types.on.store.delete

Type for store delete parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.store.get

Type for store get parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.store.list_namespaces

Type for store list namespaces parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.store.put

Type for store put parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.store.search

Type for store search parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads

Types for thread-related operations.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads.create

Type for thread creation parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads.create_run

Type for creating or streaming a run.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads.delete

Type for thread deletion parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads.read

Type for thread read parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads.search

Type for thread search parameters.

classlibs.sdk-py.langgraph_sdk.auth.types.on.threads.update

Type for thread update parameters.

classlibs.sdk-py.langgraph_sdk.cache.SWRResult

Result wrapper returned by :func:`swr`.

methodlibs.sdk-py.langgraph_sdk.cache.SWRResult.mutate(value:T=...) -> T

Update or revalidate the cached value.

funclibs.sdk-py.langgraph_sdk.cache.cache_get(key:str) -> Any | None

Get a value from the cache.

funclibs.sdk-py.langgraph_sdk.cache.cache_set(key:str, value:Any, *ttl:timedelta | None=None) -> None

Set a value in the cache.

methodlibs.sdk-py.langgraph_sdk.encryption.Encryption.get_json_decryptor(_model:str | None=None) -> types.JsonDecryptor | None

Get the JSON decryptor.

methodlibs.sdk-py.langgraph_sdk.encryption.Encryption.get_json_encryptor(_model:str | None=None) -> types.JsonEncryptor | None

Get the JSON encryptor.

classlibs.sdk-py.langgraph_sdk.encryption.LangGraphBetaWarning

Warning for beta features in LangGraph SDK.

classlibs.sdk-py.langgraph_sdk.encryption.types.EncryptionContext

Context passed to encryption/decryption handlers.

classlibs.sdk-py.langgraph_sdk.schema.Assistant

Represents an assistant with additional properties.

classlibs.sdk-py.langgraph_sdk.schema.AssistantBase

Base model for an assistant.

classlibs.sdk-py.langgraph_sdk.schema.AssistantVersion

Represents a specific version of an assistant.

classlibs.sdk-py.langgraph_sdk.schema.AssistantsSearchResponse

Paginated response for assistant search results.

classlibs.sdk-py.langgraph_sdk.schema.Checkpoint

Represents a checkpoint in the execution process.

classlibs.sdk-py.langgraph_sdk.schema.CheckpointPayload

Payload for a checkpoint event.

classlibs.sdk-py.langgraph_sdk.schema.CheckpointTaskPayload

A task entry within a `CheckpointPayload`.

classlibs.sdk-py.langgraph_sdk.schema.Config

Configuration options for a call.

classlibs.sdk-py.langgraph_sdk.schema.Cron

Represents a scheduled task.

classlibs.sdk-py.langgraph_sdk.schema.CronUpdate

Payload for updating a cron job.

classlibs.sdk-py.langgraph_sdk.schema.CustomStreamPart

Stream part emitted for `stream_mode="custom"`.

classlibs.sdk-py.langgraph_sdk.schema.DebugStreamPart

Stream part emitted for `stream_mode="debug"`.

classlibs.sdk-py.langgraph_sdk.schema.GraphSchema

Defines the structure and properties of a graph.

classlibs.sdk-py.langgraph_sdk.schema.Interrupt

Represents an interruption in the execution flow.

classlibs.sdk-py.langgraph_sdk.schema.LangSmithTracing

Configuration for LangSmith tracing.

classlibs.sdk-py.langgraph_sdk.schema.ListNamespaceResponse

Response structure for listing namespaces.

classlibs.sdk-py.langgraph_sdk.schema.MetadataStreamPart

Control event with `run_id` and other run metadata.

classlibs.sdk-py.langgraph_sdk.schema.Run

Represents a single execution run.

classlibs.sdk-py.langgraph_sdk.schema.RunCreate

Defines the parameters for initiating a background run.

classlibs.sdk-py.langgraph_sdk.schema.RunCreateMetadata

Metadata for a run creation request.

classlibs.sdk-py.langgraph_sdk.schema.RunMetadataPayload

Payload for the `metadata` control event.

classlibs.sdk-py.langgraph_sdk.schema.SearchItemsResponse

Response structure for searching items.

classlibs.sdk-py.langgraph_sdk.schema.StreamPart

Represents a part of a stream response.

classlibs.sdk-py.langgraph_sdk.schema.TaskPayload

Payload for a task start event.

classlibs.sdk-py.langgraph_sdk.schema.TaskResultPayload

Payload for a task result event.

classlibs.sdk-py.langgraph_sdk.schema.TasksStreamPart

Stream part emitted for `stream_mode="tasks"`.

classlibs.sdk-py.langgraph_sdk.schema.Thread

Represents a conversation thread.

classlibs.sdk-py.langgraph_sdk.schema.ThreadState

Represents the state of a thread.

classlibs.sdk-py.langgraph_sdk.schema.ThreadTask

Represents a task within a thread.

classlibs.sdk-py.langgraph_sdk.schema.UpdatesStreamPart

Stream part emitted for `stream_mode="updates"`.

classlibs.sdk-py.langgraph_sdk.schema.ValuesStreamPart

Stream part emitted for `stream_mode="values"`.

classlibs.sdk-py.langgraph_sdk.sse.BytesLineDecoder

Handles incrementally reading lines from text.

funclibs.sdk-py.langgraph_sdk.stream.subscription.filter_covers(coverer:dict[str, Any], target:dict[str, Any]) -> bool

Whether `coverer` is a superset of `target`.

funclibs.sdk-py.langgraph_sdk.stream.subscription.is_prefix_match(event_namespace:Namespace, prefix:Namespace) -> bool

Whether `event_namespace` starts with `prefix`.

funclibs.sdk-py.langgraph_sdk.stream.subscription.normalize_segment(segment:str) -> str

Strip the dynamic suffix after `:` from a namespace segment.

classlibs.sdk-py.langgraph_sdk.stream.transport.base.EventStreamHandle

Handle for one async filtered event stream.

classlibs.sdk-py.langgraph_sdk.stream.transport.base.SyncEventStreamHandle

Handle for one sync filtered event stream.

About this data

These signatures were extracted from the public source of langchain-ai/langgraph 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