[BUG] Rate limiter not functioning correctly "API Error: Rate limit reached" when spawning more that one agent

Found 1 possible duplicate issue:

1. https://github.com/anthropics/claude-code/issues/38123

This issue will be automatically closed as a duplicate in 3 days.

• If your issue is a duplicate, please close it and 👍 the existing issue instead
• To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

This also happens without an agent spawning I just restarted and now at 2.1.86 with the same issue.

<img width="838" height="549" alt="Image" src="https://github.com/user-attachments/assets/df1b2e18-9c78-47cc-937e-9b7766d67f55" />

I'm getting API Error: Rate limit reached even when I'm not spawning a second agent. I have Max 20x and I was using it the first time today and I immediately got API Error: Rate limit reached ... This is frustrating

<img width="1018" height="584" alt="Image" src="https://github.com/user-attachments/assets/993bfec3-43e3-4ef1-867e-927f5ff66154" />

Still broken. Where is the response from Anthropic on this critical bug?

I'm getting this as well.

It's frustrating there's no status update -- are we just a small minority getting this error? "All Systems Operational" is not right.

@andrewroth @devchris @rwp65 I had the same issue and was even getting charged additional extra usage fees while having 99% of my 20xMax Plan available after weekly reset. For whatever reason, the only thing that worked was using the /model command to switch my model and it went back to using my actual plan.

@newcomerd33-wq - what model were you using and what model did you switch to? I have found Opus is a much better model for our work and I don't want to change. However, this limit issue makes Claude Code unusable. Did you try to switch back to see if it is still an issue?

No amount of switching models helps me.

Anthropic - any response to this issue. it is still a bug.

My Claude says your issue shares the same rate limit pattern as #38239, #37917, #39465, #39938, #40079, #40085, #40128. Your case adds a specific trigger: multi-agent spawning. Linking so Anthropic sees the full scope of this rate limit problem.

A user in #40584 may have found the root cause for your rate limit issue. Two findings:

1. The "Rate limit reached" error is generated CLIENT-SIDE by Claude Code, not by the Anthropic API. Evidence: model: "<synthetic>", input_tokens: 0, output_tokens: 0 — no request was ever sent to the API. The client-side rate limiter counts the entire context size against the token budget, so large/long sessions get permanently blocked even when the API is fully available.
2. The client-side rate limiter MULTIPLIES estimated context tokens by the number of concurrent requests. So 3 subagents × 400K context = 1.2M tokens counted against budget — even though the context is cached and shared. This means as your session grows, fewer concurrent agents can run before hitting the false limit.

See #40584 for the full analysis with log evidence from two independent occurrences.

WHERE IS THE RESPONSE FROM ANTHROPIC????????????

A user in #40584 may have found the root cause for your rate limit issue. Two findings: 1. The "Rate limit reached" error is generated CLIENT-SIDE by Claude Code, not by the Anthropic API. Evidence: model: "", input_tokens: 0, output_tokens: 0 — no request was ever sent to the API. The client-side rate limiter counts the entire context size against the token budget, so large/long sessions get permanently blocked even when the API is fully available. 2. The client-side rate limiter MULTIPLIES estimated context tokens by the number of concurrent requests. So 3 subagents × 400K context = 1.2M tokens counted against budget — even though the context is cached and shared. This means as your session grows, fewer concurrent agents can run before hitting the false limit. See #40584 for the full analysis with log evidence from two independent occurrences.

I could somewhat verify cause 1 from my experience. existing long sessions got hit with rate limit error, but new sessions worked fine.

Closing for now — inactive for too long. Please open a new issue if this is still relevant.

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.