Models
Connect model providers and choose the model each agent runs on.
Lyra does not ship its own model. You connect the providers you want to use, and each agent runs on the model you pick for it.
Supported providers
- Anthropic
- OpenAI
- Any OpenAI-compatible endpoint, including self-hosted inference servers and gateways
Connect a provider
Open your space settings, add a provider, paste its API key and test the connection. Keys are encrypted at rest and are never sent to the browser.
Choose a model per agent
When you create an agent, pick one of the models from your connected providers. Different agents in the same workspace can use different models, so a research agent and a coding agent can each run on what suits them.