OpenCode is an open-source terminal AI coding agent. It includes routing.run as a login provider, so you can authenticate with a dashboardDocumentation Index
Fetch the complete documentation index at: https://docs.routing.run/llms.txt
Use this file to discover all available pages before exploring further.
rk_ key and use any model id that starts with route/.
Official references:
Install OpenCode
Pick one install path, then confirm the binary is on yourPATH:
opencode --help (or opencode) from a project directory.
Connect routing.run
Get your API key
Create an API key at app.routing.run. Copy the full secret — it starts with
rk_.Login command
opencode.json for the standard routing.run setup.
If list calls work but chat fails, confirm the model id includes the
route/ prefix and matches a published routing.run model. In OpenCode model selectors and -m, use the full provider/model form, for example routing.run/route/kimi-k2.6.Verify
After logging in, you can verify that the endpoint is reachable:Troubleshooting
- Wrong host or 401 — Run
opencode auth login, select routing.run, and enter a validrk_key from the dashboard. - Unknown model — Use ids from the models catalog or dashboard; typos in
route/…ids returnINVALID_MODELfrom the API. - Credential not found — Run
opencode auth listto check if the routing.run credential is stored. If not, runopencode auth loginagain.
Recommended models
Published model IDs are listed on the models page. Exact access depends on your plan tier and the dashboard; common picks:| Model | Use case |
|---|---|
route/step-3.5-flash | Best StepFun model for fast agent work |
route/step-3.5-flash-2603 | StepFun fallback or alternate release |
route/kimi-k2.6 | Best default for agentic and tool-heavy flows |
route/kimi-k2.6-precision | Higher-precision Kimi choice |
route/glm-5.1 | Strong GLM default |
route/glm-5.1-precision | Strongest GLM reasoning-first choice |