Claude Desktop and AI Hub
Connect Claude Desktop to AI Hub's Messages API: where the setting lives, the base URL and the model IDs.
Connect
AI Hub serves the Anthropic Messages API at https://tokenhub-api.belive.sg/api. Enter that base, a workspace inference key and AI Hub model IDs in Desktop's third-party inference form. The CLI setup is in the Claude Code guide.
Anthropic's Desktop gateway requirements describe the fields the form takes.
Where desktop configuration lives
Anthropic documents a local setup form under Help → Troubleshooting → Enable Developer Mode, then Developer → Configure Third-Party Inference after restart. An administrator-managed configuration can take precedence and make the form read-only. See the desktop connection instructions.
The form takes a gateway URL, its authentication settings and model mappings. Enter the Messages base https://tokenhub-api.belive.sg/api, not the Chat Completions base. Desktop reads third-party inference configuration, not the CLI's ANTHROPIC_BASE_URL or settings.json.
Models and keys
Choose exact author/slug IDs from the model catalogue for the primary and auxiliary model selections, and add any fallback routes you want. Create a dedicated workspace inference key in the same environment as the API, keep it in a secret store, and set a spend cap. An MCP connector provides tools; the model inference provider is set in the form.
curl --fail "https://tokenhub-api.belive.sg/api/v1/models"Troubleshooting
- Gateway unreachable: verify the configured gateway host and network path.
- A Messages endpoint 404: verify the environment and exact path; AI Hub serves
/api/v1/messages. - CLI settings have no effect: use the Desktop third-party inference configuration and check whether administrator policy controls it.
- Check Activity for the model, usage and cost of each request.
Other clients
For a Chat Completions setup, use the Qwen Code, Hermes Agent, OpenClaw or OpenCode guide. Do not put credentials in repository config, screenshots or diagnostic logs.