OLO

Connect models and tools in repeatable workflows.

OLO, the Open LLM Orchestrator, coordinates language model calls, tools and workflow steps. Define workflows as portable JSON graphs and run them on Temporal workers, with model endpoints and deployment settings configured separately.

Apache-2.0 Java 21, Temporal

Take a closer look

Diagram of an OLO workflow graph from start node to return value
Diagram: a workflow is a portable graph of LLM, tool and hook nodes.
Diagram of a chat message flowing through the OLO backend, Temporal, the worker and the kernel
Diagram: how a chat message becomes a workflow run on Temporal.

Diagram: a workflow is a portable graph of LLM, tool and hook nodes.

What you can do

  • Choose hosted or local modelsConfigure compatible model endpoints separately from the workflow graph. Evaluate each model for the tools and outputs your workflow needs.
  • Portable workflow graphsDefine steps, connections, variables and return values in versioned JSON. Use builders and validation to maintain your workflows.
  • Start with a presetActivate a ready-made workflow from Administration, Scenarios in the OLO console.
  • Run on Temporal workersWorkers execute workflow steps and report progress back to the chat interface.
  • Extend your workflowsAdd custom nodes, tools and hooks through the service provider interface (SPI), and expose them to workflow editors.

Explore the demos

Choose a demo below to see what it includes and how to get started.

Install & Deploy

Run the OLO stack with Docker Compose or build a Temporal worker from source. Kubernetes and Helm installation instructions are not currently provided in the OLO deployment repository.

Developer resources and contributing

Help improve the project through code, documentation, testing or bug reports. Start with the contribution guide and find an issue that matches your interests.

Marketplace

OLO Marketplace Browse checked and signed ToolGate packages. Review included tools, settings and compatibility before installing.