You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
简化第三方模型配置,增加预制模板Simplify third-party model config with pre‑built templates #5350
Problem: Configuring third‑party compatible providers (OpenCode Zen, OpenCode Go, Agnes, Meituan Sensenova) requires manually entering Base URL, model names, and API key env vars with no in‑UI docs. After saving, model lists often stuck at not checked / cache failed and won't load.
Suggestion: Add built‑in templates (pre‑filled URLs & common model lists) – users only input their API key. Embed official docs/parameters on the config page. Add a "Test Connection" button to verify and auto‑update status. Fix the cache failure bug.
Benefit: New users can finish setup within 1 minute, without external searches or lengthy debugging.
This issue is staying open for maintainer triage. CodeWhale gets better because people bring us real edge cases from real machines, providers, regions, and workflows.
If you can add a reproduction, logs, version output, screenshots, or the provider/model involved, that makes it much easier for us to verify and harvest the fix. Maintainers may comment /lgtmi to mark recurring issue reporters as approved so this intake note is skipped next time.
Problem: Configuring third‑party compatible providers (OpenCode Zen, OpenCode Go, Agnes, Meituan Sensenova) requires manually entering Base URL, model names, and API key env vars with no in‑UI docs. After saving, model lists often stuck at not checked / cache failed and won't load.
Suggestion: Add built‑in templates (pre‑filled URLs & common model lists) – users only input their API key. Embed official docs/parameters on the config page. Add a "Test Connection" button to verify and auto‑update status. Fix the cache failure bug.
Benefit: New users can finish setup within 1 minute, without external searches or lengthy debugging.
So I think I actually simplified it a bit too much - working on fine tuning this!
Activity
github-actions commentedon Aug 13, 2026
Thanks @shadapang for the report.
This issue is staying open for maintainer triage. CodeWhale gets better because people bring us real edge cases from real machines, providers, regions, and workflows.
If you can add a reproduction, logs, version output, screenshots, or the provider/model involved, that makes it much easier for us to verify and harvest the fix. Maintainers may comment
/lgtmito mark recurring issue reporters as approved so this intake note is skipped next time.Hmbown commentedon Aug 14, 2026
Moved to v0.9.9. Truth-boundary (/models 2xx ≠ model ready) is on main. Prefab third-party templates are later UX.
Hmbown commentedon Aug 15, 2026
ah - thank you so much fo rthis - I 'mg oing to try to move it up to 0.9.8
fix(tui): do not mark provider-id helper as a unit test
feat(tui): add a SenseNova OpenAI-compatible setup preset
docs: record the rest of the 0.9.8 assigned finish
Hmbown commentedon Aug 17, 2026
Rolled to v0.9.10: the v0.9.9 release branch merged to main tonight (#5476) without this landing; nothing here is closed or dropped.
feat(tui): add beginner provider setup templates (#5350)
Hmbown commentedon Aug 23, 2026
So I think I actually simplified it a bit too much - working on fine tuning this!
Hmbown commentedon Aug 23, 2026
Verified shipped on main during 0.9.12 triage. Evidence: a0f5aa5 added beginner provider setup templates (#5350).
shadapang commentedon Aug 29, 2026
@Hmbown 非常感谢看到了我的需求,我也已经尝试过了,比之前体验好了很多,谢谢
docs(providers): every supported host appears in PROVIDERS.md, enforced