anomalyco / anomalyco/opencode
Frequent rate_limit_exceeded on Muse Spark 1.3 Contributor with paid OpenCode Go
@fwang is already working on this.
Since Sep 8, 2026.
- Dominant language
- TypeScript
- Stars
- 209k
- Forks
- 27.5k
- PR merge metrics
- PR metrics pending
Description
Description
I'm using a paid OpenCode Go subscription and Muse Spark 1.3 Contributor keeps failing with:
Upstream request failed: [rate_limit_exceeded]
This happens pretty often during normal coding sessions. I'm not sending a huge number of requests, and my OpenCode Go quota is nowhere near exhausted.
Other models on OpenCode Go continue working, so this looks like a Muse-specific or upstream capacity limit rather than my subscription quota.
The main problem is that Muse becomes pretty unreliable for agent work because a request can suddenly fail in the middle of a task.
Since Go is a paid subscription, I'd like to understand what exactly is being limited here.
Is there a separate rate or capacity limit for Muse that isn't reflected in the OpenCode Go usage quota?
And if this is an upstream capacity issue, could OpenCode retry these requests automatically with some backoff instead of immediately failing the agent run?
I've been seeing this repeatedly on September 7-8, 2026.
Is this a known issue?
Plugins
No response
OpenCode version
1.18.29
Steps to reproduce
Start OpenCode with an active paid OpenCode Go subscription.
Select Muse Spark 1.3 Contributor.
Use it normally for coding/agent work.
After some requests, Muse starts failing with:
Upstream request failed: [rate_limit_exceeded]
This happens repeatedly during normal interactive usage. My OpenCode Go quota is still available when it happens, and other OpenCode Go models continue to work.
There doesn't seem to be any unusual request burst or other obvious trigger — it can happen in the middle of an ordinary agent session.
Screenshot and/or share link
No response
Operating System
macOS 15.7.7, Apple Silicon (M1 Pro)
Terminal
iterm2
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.