[BUG] Usage limit reached despite UI showing low consumption percentages

Resolved 💬 6 comments Opened Apr 20, 2026 by yarivh54 Closed May 30, 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?

Summary

Claude Code displays "Usage limit reached" while the Plan Usage UI simultaneously shows very low consumption across all tracked metrics. The displayed percentages are clearly inconsistent with the actual enforcement happening under the hood.

What's Wrong?

The Claude Code interface blocked all usage (both Opus 4.7 and Sonnet 4.6) while the Plan Usage panel showed:

  • 5-hour session limit: ~1% used
  • Weekly · all models: ~17% used
  • Weekly · Claude Design (Opus): ~34% used
  • Weekly · Sonnet only: ~17% used

None of these values are near 100%, yet the system returned a hard block with "Usage limit reached" and an Upgrade prompt. Switching from Opus 4.7 to Sonnet 4.6 did not resolve the issue — confirming this is not a model-specific limit.

Actual Behavior

Hard block on all usage. UI percentages appear to be decorative rather than accurately reflecting actual enforcement state.

What Should Happen?

Usage limits should be enforced consistently with what the UI displays. If the 5-hour session is at 1% and weekly limits are at 17%, the user should not be blocked.

Error Messages/Logs

Steps to Reproduce

  1. Open Claude Code (desktop app)
  2. Check Plan Usage panel — all metrics show low percentages
  3. Attempt to send any message (any model)
  4. Receive "Usage limit reached" block despite UI showing available headroom

Claude Model

Sonnet (default)

Is this a regression?

Yes, this worked in a previous version

Last Working Version

Sonnet 4.6

Claude Code Version

Latest

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Warp

Additional Information

This makes it impossible to plan or manage usage effectively. If the displayed metrics cannot be trusted, the entire usage tracking UI is misleading paying subscribers. Related to #11810 and #9094 — this issue appears to still be actively affecting users.

View original on GitHub ↗

This issue has 6 comments on GitHub. Read the full discussion on GitHub ↗