Skip to main content
omp connects directly to LithosAI through its OpenAI Chat Completions transport. You do not need a local proxy or protocol adapter.

Before you begin

Install omp, create a key on the API Keys page, and confirm the model ID and token limits on the Models page.
1

Export your LithosAI key

Export the key in the shell where you start omp.
2

Add LithosAI as a custom provider

Merge the following provider into ~/.omp/agent/models.yml.
~/.omp/agent/models.yml
apiKey names the environment variable that contains the credential. Replace the model ID, context window, and output limit with the values shown in the LithosAI console.
3

Confirm and select the model

Confirm that omp loaded the provider, then start a session and select the model.
Inside the session, open the model picker and choose lithosai/moonshotai/Kimi-K3.
omp evolves quickly. If models.yml fails validation, compare your installed version with the current provider schema.

Verify the connection

From a project that contains a README.md, ask omp:
omp should read the local file and return its heading. This verifies both the model response and a basic local tool call.

Troubleshooting

Issues common to every agent are listed under Common issues.

Resources