AI Potluck
Product / UX / Orchestration & agents

LangGraph

LangChain

Low-level orchestration framework from LangChain for stateful, multi-actor agent applications. It exposes an explicit graph and state-machine model - nodes, edges, persistence and human-in-the-loop checkpoints - rather than a higher-level abstraction, which is what makes fine-grained control over loops, branching and durable resumable execution possible.

The LangGraph Platform and LangSmith are the separately sold managed layer. Verified 2026-08-11 via the langchain-ai/langgraph repository.

Openness

4 high confidence
4.0
license
MIT(core)
commercial
LangGraph-Platform/LangSmith
source
public
core-gated
gated(the Agent Server runtime ships as the closed Elastic-2.0 langgraph-api package and self-hosting it needs LANGGRAPH_CLOUD_LICENSE_KEY)

MIT-licensed self-hostable core with a commercial managed deployment platform beside it, and the gating is real rather than just a consequence of that platform existing. A full untruncated recursive tree of langchain-ai/langgraph at main is 830 entries with nine LICENSE files, all per-package copies of the MIT root, and no enterprise or ee directory, so the gate is not inside the repo. It is that a piece of the runtime is not in the repo at all: PyPI reports langgraph-api, the Agent Server, under Elastic-2.0 while langgraph itself is MIT, no langgraph-api source repo exists under the langchain-ai organization, and LangChain's own self-host documentation requires a LangGraph Cloud license key plus egress to beacon.langchain.com for license verification. A license key you must hold to run the server is functionality withheld from the published source for a paid tier. That is what separates LangGraph from LangChain, which publishes everything it ships.

Adoption

5 high confidence
5.0

72,717,229 PyPI downloads in the trailing 30 days for langgraph. Bands at level 5 (>10M) on the software and model adoption scale. Largely transitive via LangChain, so directional.

Capability

5 high confidence
5.0

Deepest durable-execution control among the orchestration peers.

  • https://github.com/langchain-ai/langgraph recorded 2026-08-13

    README - durable execution resuming from where a run left off, human-in-the-loop state inspection and modification, streaming, memory and persistence, branching and subgraphs; 39.6k stars, 6.7k forks

Verified 2026-08-11