Run LangGraph agents on a governed runtime — Kohala
Governed runs for your LangGraph state machines. Import your graph or wrap it with the kohala SDK — charters, billing caps, credit metering, and trace IDs on every invoke, with LangGraph still managing state.
Explore Kohala
Frequently Asked Questions
- Can I run a LangGraph graph on Kohala?
- Yes. Import your LangGraph code through the wizard and it runs on Kohala's governed runtime under a recorded charter, a server-enforced allowlist on the Kohala tools it can call, and a billing cap. Or wrap your compiled graph with the kohala SDK, keep it on your own infrastructure, and report each run — recorded under the same charter, held to the same caps and validators, metered and traced the same way. Unless you ask for a cadence when you approve it, an imported graph arrives unscheduled — it runs when you trigger it.
- Does Kohala replace LangGraph's state management?
- No. LangGraph keeps managing your graph state and control flow, including branches that form decision trees. Kohala adds the layer around it: governance policies, cost metering, output validators, and an account-scoped shared metered memory API your graph can read and write across runs once you wire it in. Each connected graph reports to one Kohala account; workflows never span accounts.
- How is a governed LangGraph run billed?
- Runs are metered in platform credits against a per-agent billing cap you set. Each run reports its exact credit cost, latency, and a trace ID, so you always know what a graph execution cost.