Available models
List public models available to your account and its current access rules.
GET /v1/models Compatibility API
Your application. A familiar API.
Use familiar OpenAI and Anthropic request formats to connect your application to available Lurus models. Keep model access and billing in one account, with transparent protocol limits.
Requires a Lurus API key, an available model and sufficient credit. Protocol and field support depend on the selected model.
Four endpoints. Clear responsibilities.
Chat Completions
Use the OpenAI chat request format, with complete or streaming responses.
POST /v1/chat/completions Responses
Use the Responses protocol within the capabilities of the selected route.
POST /v1/responses Messages
Use Anthropic-style Messages requests with model-dependent support.
POST /v1/messages Configure, then verify
A familiar protocol is not universal compatibility.
Use https://backend.lurus.sh/v1 as the client base URL and authenticate with a Lurus key via Bearer or x-api-key. Test the fields, streaming and tool-call format your application needs against the chosen model.
Connect on the server side.
- Create a named API key in API access.
- Store it in a secret manager or server environment, never a browser bundle.
- Choose an available model and verify a minimal request before production use.
Model inference, not the whole coding agent.
The gateway returns model responses and tool-call data. It does not run your application tools or provide the CLI’s complete coding-agent runtime. Your application remains responsible for its tool execution and authorisation.
Your code stays yours.
Source code and request context are processed with Zero Data Retention. No use for model training. Providers and processing regions are disclosed transparently.
See privacy and processing detailsA few things to know before you start.
Which protocols are available?
The gateway exposes OpenAI Chat Completions, OpenAI Responses and Anthropic Messages, plus a model list. Capabilities and accepted fields depend on the selected model route.
Does the API run tools for my application?
No. Model responses can contain tool-call data. Your application handles its own tool execution, permissions and results; the compatibility gateway is not the full CLI agent.
Is usage included without limits?
No. Model access depends on the account, trial and team rules. Requests require sufficient credit and are subject to usage and rate limits.
Compatibility API
Connect the next application deliberately.
Start in API access, choose a model and verify the request format your application actually needs.
Requires a Lurus API key, an available model and sufficient credit. Protocol and field support depend on the selected model.
Want to start with your team? Talk to us.