Documentation
Inert is an LLM gateway that speaks the Anthropic Messages API (/v1/messages), the OpenAI Chat Completions API (/v1/chat/completions) and the OpenAI Responses API (/v1/responses). Any harness that speaks one of them can point at it by changing a base URL and a key. The base URL is https://api.inert.lol, authenticated with either an x-api-key or an Authorization: Bearer header. Create an account, generate a key in the dashboard, then send a request. Keys are shown once, so store one like a secret. The model IDs are the route names, kebab-cased: fable-5 and opus-5. GET /v1/models is always the live list.
Compatibility
| Harness | Works today? | How |
|---|---|---|
| Claude Code | Yes | Native Anthropic client. Bare base URL plus an auth token. |
| OpenCode | Yes | Provider override in opencode.json. |
| Kilo Code | Yes | "Anthropic Messages" custom provider. |
| Pi | Yes | Provider override in ~/.pi/agent/models.json. |
| Hermes | Yes | custom_providers entry in ~/.hermes/config.yaml. |
| Codex CLI | Yes | model_providers entry in ~/.codex/config.toml. |
| Cursor | Not yet | OpenAI override not verified against Inert yet. |
| Kiro | Not yet | Fixed model list, no BYOK or custom endpoint. Feature request open upstream. |
Next steps
- QuickstartGet a key, fund it, and make your first request.
- Claude CodeThe most common setup, start to finish.