[BUG] This is frustating to get usage limit issue because of Anthropic capacity limitation

Status Open
Maintainer reply None cached
Activity 4 comments · opened Jun 6, 2026

Preflight Checklist

  • [x] I have searched existing issues and this hasn't been reported yet
  • [x] This is a single bug report (please file separate reports for different bugs)
  • [x] I am using the latest version of Claude Code

What's Wrong?

Title: [Bug] API Error: "Server is temporarily limiting requests (not your usage limit) · Rate limited" despite low usage on Claude Max

Bug Description
I am on the Claude Max plan and repeatedly receive the error:
"API Error: Server is temporarily limiting requests (not your usage limit) · Rate limited"
while working in Claude Code (VS Code extension), even though my usage meters show plenty of headroom:

  • Session (5hr): 47%
  • Weekly (7 day): 20%
  • Weekly Sonnet: 3%

I also intermittently see a "You've hit your usage limit" banner even though none of my limits are reached, plus errors like "an image in the conversation could not be processed and was removed."

Steps to Reproduce

  1. Open Claude Code in VS Code on a Max account
  2. Continue an active coding session (mid-task edits)
  3. Requests start failing with the server-side rate limit error after a few seconds

Expected Behavior
Requests should succeed within my Max plan limits; server-side capacity issues should not block paid usage for extended periods, and the UI should not show a misleading "usage limit reached" banner.

Actual Behavior
Repeated server-side rate limiting blocks work despite low personal usage. Occurred on Saturday, June 6, 2026 (US Eastern), during my limited weekend working time.

Environment

  • Plan: Claude Max
  • Platform: Windows
  • Client: Claude Code in VS Code (integrated)
  • Auth method: Claude AI (claude.ai login)
  • Location: Florida, US

Additional Context
This appears related to the ongoing June capacity incidents (e.g., issues #52784 and #53915 report the same error). Please confirm whether failed/rate-limited requests still consume session quota, as my session meter advanced during the errors.

What Should Happen?

Title: [Bug] API Error: "Server is temporarily limiting requests (not your usage limit) · Rate limited" despite low usage on Claude Max

Bug Description
I am on the Claude Max plan and repeatedly receive the error:
"API Error: Server is temporarily limiting requests (not your usage limit) · Rate limited"
while working in Claude Code (VS Code extension), even though my usage meters show plenty of headroom:

  • Session (5hr): 47%
  • Weekly (7 day): 20%
  • Weekly Sonnet: 3%

I also intermittently see a "You've hit your usage limit" banner even though none of my limits are reached, plus errors like "an image in the conversation could not be processed and was removed."

Steps to Reproduce

  1. Open Claude Code in VS Code on a Max account
  2. Continue an active coding session (mid-task edits)
  3. Requests start failing with the server-side rate limit error after a few seconds

Expected Behavior
Requests should succeed within my Max plan limits; server-side capacity issues should not block paid usage for extended periods, and the UI should not show a misleading "usage limit reached" banner.

Actual Behavior
Repeated server-side rate limiting blocks work despite low personal usage. Occurred on Saturday, June 6, 2026 (US Eastern), during my limited weekend working time.

Environment

  • Plan: Claude Max
  • Platform: Windows
  • Client: Claude Code in VS Code (integrated)
  • Auth method: Claude AI (claude.ai login)
  • Location: Florida, US

Additional Context
This appears related to the ongoing June capacity incidents (e.g., issues #52784 and #53915 report the same error). Please confirm whether failed/rate-limited requests still consume session quota, as my session meter advanced during the errors.

Error Messages/Logs

Steps to Reproduce

Title: [Bug] API Error: "Server is temporarily limiting requests (not your usage limit) · Rate limited" despite low usage on Claude Max

Bug Description
I am on the Claude Max plan and repeatedly receive the error:
"API Error: Server is temporarily limiting requests (not your usage limit) · Rate limited"
while working in Claude Code (VS Code extension), even though my usage meters show plenty of headroom:

  • Session (5hr): 47%
  • Weekly (7 day): 20%
  • Weekly Sonnet: 3%

I also intermittently see a "You've hit your usage limit" banner even though none of my limits are reached, plus errors like "an image in the conversation could not be processed and was removed."

Steps to Reproduce

  1. Open Claude Code in VS Code on a Max account
  2. Continue an active coding session (mid-task edits)
  3. Requests start failing with the server-side rate limit error after a few seconds

Expected Behavior
Requests should succeed within my Max plan limits; server-side capacity issues should not block paid usage for extended periods, and the UI should not show a misleading "usage limit reached" banner.

Actual Behavior
Repeated server-side rate limiting blocks work despite low personal usage. Occurred on Saturday, June 6, 2026 (US Eastern), during my limited weekend working time.

Environment

  • Plan: Claude Max
  • Platform: Windows
  • Client: Claude Code in VS Code (integrated)
  • Auth method: Claude AI (claude.ai login)
  • Location: Florida, US

Additional Context
This appears related to the ongoing June capacity incidents (e.g., issues #52784 and #53915 report the same error). Please confirm whether failed/rate-limited requests still consume session quota, as my session meter advanced during the errors.

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

Opus 4.8

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Terminal.app (macOS)

Additional Information

Title: [Bug] API Error: "Server is temporarily limiting requests (not your usage limit) · Rate limited" despite low usage on Claude Max

Bug Description
I am on the Claude Max plan and repeatedly receive the error:
"API Error: Server is temporarily limiting requests (not your usage limit) · Rate limited"
while working in Claude Code (VS Code extension), even though my usage meters show plenty of headroom:

  • Session (5hr): 47%
  • Weekly (7 day): 20%
  • Weekly Sonnet: 3%

I also intermittently see a "You've hit your usage limit" banner even though none of my limits are reached, plus errors like "an image in the conversation could not be processed and was removed."

Steps to Reproduce

  1. Open Claude Code in VS Code on a Max account
  2. Continue an active coding session (mid-task edits)
  3. Requests start failing with the server-side rate limit error after a few seconds

Expected Behavior
Requests should succeed within my Max plan limits; server-side capacity issues should not block paid usage for extended periods, and the UI should not show a misleading "usage limit reached" banner.

Actual Behavior
Repeated server-side rate limiting blocks work despite low personal usage. Occurred on Saturday, June 6, 2026 (US Eastern), during my limited weekend working time.

Environment

  • Plan: Claude Max
  • Platform: Windows
  • Client: Claude Code in VS Code (integrated)
  • Auth method: Claude AI (claude.ai login)
  • Location: Florida, US

Additional Context
This appears related to the ongoing June capacity incidents (e.g., issues #52784 and #53915 report the same error). Please confirm whether failed/rate-limited requests still consume session quota, as my session meter advanced during the errors.

View original on GitHub ↗

4 Comments

github-actions[bot] · 2 months ago

Found 3 possible duplicate issues:

  1. https://github.com/anthropics/claude-code/issues/53915
  2. https://github.com/anthropics/claude-code/issues/61529
  3. https://github.com/anthropics/claude-code/issues/40128

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

yurukusa · 2 months ago

A few of the things you're seeing are actually three separate failure modes, and only one of them is "you" — so it's worth pulling them apart:
1. "Server is temporarily limiting requests (not your usage limit) · Rate limited" — this is server-side overload, not your plan. The public API surfaces this as HTTP 529 overloaded_error ("The API is temporarily overloaded"), which the docs explicitly describe as happening "when APIs experience high traffic across all users." That's exactly why the message says not your usage limit — and your 47% / 20% / 3% meters confirm you have headroom. Nothing client-side fixes a 529; it clears when capacity frees up, so back-off-and-retry is the only real lever. (Source: https://platform.claude.com/docs/en/api/errors)
There's one adjacent edge the same docs call out: a sharp spike in your own traffic can trip 429 rate_limit_error via "acceleration limits," and the fix there is to ramp up gradually instead of bursting. So a quick way to tell which one you're hitting: if failures cluster around heavy parallel / sub-agent fan-out, suspect the 429 acceleration path; if they happen during steady single-threaded editing, it's the 529 capacity path and genuinely not on you.
2. The "You've hit your usage limit" banner while your meters show headroom — that's a display mislabel, not a real limit. A 529 (server capacity) is being rendered with usage-limit wording. Your own meters are the authoritative source, and they say you're fine. Same root as the threads you linked (#52784 / #53915).
3. "an image in the conversation could not be processed and was removed" — unrelated to the rate limiting. That's image handling (an oversized or unsupported image in context); very large images can also hit the 32 MB request cap (413 request_too_large). Worth filing separately if it reproduces.
On your actual question — "do failed/rate-limited requests still consume session quota?" I won't guess at Anthropic's billing internals, but you can measure it cleanly: note /usage, trigger the error, check /usage again. A 529 that fails before any tokens are generated shouldn't bill tokens; if your meter still advances across a pure-error window, that's a concrete bug — and the single most useful thing to attach is a request-id. Run with --debug and the request-id header is printed to stderr; that lets support trace the exact failing request.
TL;DR: #1 is server-side and not your fault, the banner in #2 is just mislabeling it, and #3 is a separate image issue. The only thing on your side worth ruling out is whether a traffic burst is tripping 429 acceleration limits versus steady-state 529s.

wozcode-helper · 2 months ago

i hit the same "server is temporarily limiting requests" in claude code on claude max (session meter was ~47% yet requests kept failing saturday while i was in vscode). tbh wozcode (a plugin on top of claude code) cut ur token spend ~50% w better caching — https://wozcode.com

kolsha · 1 month ago

<img width="784" height="543" alt="Image" src="https://github.com/user-attachments/assets/6c639986-e7c4-4c2f-b3e7-764f6fe425f1" />

Clearly I did not use claude today, but still my 5-hour window is exhausted. How come, @anthropics?