[FEATURE]: AutoFallback model #3891

Closed
opened 2026-02-16 17:41:50 -05:00 by yindo · 1 comment
Owner

Originally created by @riccardodebellini on GitHub (Dec 26, 2025).

Originally assigned to: @thdxr on GitHub.

Feature hasn't been suggested before.

  • I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

It would be interesting to have a feature whereby when a model fails (especially errors due to exceeding the limit of requests/tokens per minute/hour/day), a switch to another is automatically made, always set by the user to continue the flow of thoughts/actions.
This would be useful because many providers, such as Google AI Studio, count as requests not only the initial user input, but also the result of executing a command, reading a file, etc., quickly reaching limits, especially for users on free plans (Gemini 3.0 Flash offers 20 RPD).


Example:

  • Starts with Gemini 3.0 Flash - 20 requests
  • Falls back to Gemini 2.5 Flash - 20 requests
  • Falls back to Openrouter/Deepseek - 20 requests
  • Uses free model (Big Pickle or Groq through Opencode Zen)
Originally created by @riccardodebellini on GitHub (Dec 26, 2025). Originally assigned to: @thdxr on GitHub. ### Feature hasn't been suggested before. - [x] I have verified this feature I'm about to request hasn't been suggested before. ### Describe the enhancement you want to request It would be interesting to have a feature whereby when a model fails (especially errors due to exceeding the limit of requests/tokens per minute/hour/day), a switch to another is automatically made, always set by the user to continue the flow of thoughts/actions. This would be useful because many providers, such as Google AI Studio, count as requests not only the initial user input, but also the result of executing a command, reading a file, etc., quickly reaching limits, especially for users on free plans (Gemini 3.0 Flash offers 20 RPD). --- Example: - Starts with Gemini 3.0 Flash - 20 requests - Falls back to Gemini 2.5 Flash - 20 requests - Falls back to Openrouter/Deepseek - 20 requests - Uses free model (Big Pickle or Groq through Opencode Zen)
yindo added the discussionzen labels 2026-02-16 17:41:50 -05:00
yindo closed this issue 2026-02-16 17:41:50 -05:00
Author
Owner

@github-actions[bot] commented on GitHub (Dec 26, 2025):

This issue might be a duplicate of existing issues. Please check:

  • #1267: Support per rule model fallbacks for outages and credit depletion
  • #1326: Support multiple API keys per provider with fallback logic (for handling rate limits)
  • #5748: Proposal: Provider Auth v2 (encrypted credential vault + multi-account OAuth + same-request rotation)

Feel free to ignore if none of these address your specific case.

@github-actions[bot] commented on GitHub (Dec 26, 2025): This issue might be a duplicate of existing issues. Please check: - #1267: Support per rule model fallbacks for outages and credit depletion - #1326: Support multiple API keys per provider with fallback logic (for handling rate limits) - #5748: Proposal: Provider Auth v2 (encrypted credential vault + multi-account OAuth + same-request rotation) Feel free to ignore if none of these address your specific case.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#3891