Config

Optional. Python version in 'major.minor' format (e.g. '3.11'). Must be at least 3.11 or greater for this deployment to function properly.

Optional. Node.js version as a major version (e.g. '20'), if your deployment needs Node. Must be >= 20 if provided.

Optional. Base image to use for the LangGraph API server.

Defaults to langchain/langgraph-api or langchain/langgraphjs-api.

Optional. Linux distribution for the base image.

Must be one of 'wolfi', 'debian', or 'bookworm'. If omitted, defaults to 'debian' ('latest').

Optional. Path to a pip config file (e.g., "/etc/pip.conf" or "pip.ini") for controlling package installation (custom indices, credentials, etc.).

Only relevant if Python dependencies are installed via pip. If omitted, default pip settings are used.

Optional. Python package installer to use ('auto', 'pip', or 'uv').

• 'auto' (default): Use uv for supported base images, otherwise pip
• 'pip': Force use of pip regardless of base image support
• 'uv': Force use of uv (will fail if base image doesn't support it)

Optional. Explicit deployment source configuration.

Use { "kind": "uv", "root": "." } to deploy from a uv project rooted at root/pyproject.toml and root/uv.lock. If root is a workspace and the target is ambiguous, set package to the desired workspace member.

Optional. Additional Docker instructions that will be appended to your base Dockerfile.

Useful for installing OS packages, setting environment variables, etc. Example: dockerfile_lines=[ "RUN apt-get update && apt-get install -y libmagic-dev", "ENV MY_CUSTOM_VAR=hello_world" ]

List of Python dependencies to install, either from PyPI or local paths.

Optional. Named definitions of graphs, each pointing to a Python object.

Graphs can be StateGraph, @entrypoint, or any other Pregel object OR they can point to (async) context managers that accept a single configuration argument (of type RunnableConfig) and return a pregel object (instance of Stategraph, etc.).

Keys are graph names, values are either "path/to/file.py:object_name" strings or objects with a "path" key and optional "description" key. Example: { "mygraph": "graphs/my_graph.py:graph_definition", "anothergraph": { "path": "graphs/another.py:get_graph", "description": "A graph that does X" } }

Optional. Environment variables to set for your deployment.

• If given as a dict, keys are variable names and values are their values.
• If given as a string, it must be a path to a file containing lines in KEY=VALUE format.

Optional. Configuration for the built-in long-term memory store, including semantic search indexing.

If omitted, no vector index is set up (the object store will still be present, however).

Optional. Configuration for the built-in checkpointer, which handles checkpointing of state.

If omitted, no checkpointer is set up (the object store will still be present, however).

Optional. Custom authentication config, including the path to your Python auth logic and the OpenAPI security definitions it uses.

Optional. Custom at-rest encryption config, including the path to your Python encryption logic.

Allows you to implement custom encryption for sensitive data stored in the database.

Optional. Configuration for the built-in HTTP server, controlling which custom routes are exposed and how cross-origin requests are handled.

Optional. Webhooks configuration for outbound event delivery.

Forwarded into the container as LANGGRAPH_WEBHOOKS. See WebhooksConfig for URL policy and header templating details.

Optional. Named definitions of UI components emitted by the agent, each pointing to a JS/TS file.

Optional. Control whether to retain Python packaging tools in the final image.

Allowed tools are: "pip", "setuptools", "wheel". You can also set to true to include all packaging tools.