[BUG] Max x20 plan — 5-hour quota consumed in ~10 prompts (started March 14, 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?
What's Wrong?
Starting today (March 14, 2026), the 5-hour rolling quota on my Max x20 ($200/month) subscriptions is being consumed at an abnormally fast rate. The quota is exhausted after approximately 10 prompts, which is drastically fewer than expected for this tier.
Setup & Context
I have three separate Max x20 subscriptions:
Subscription A & B: Used both on remote agents (headless/distant machines) and in local terminal sessions on my PC.
Subscription C: Used exclusively in local terminal sessions on my PC — no remote agents, no shared usage.
The Problem
To isolate the issue, I tested Subscription C (local-only, no remote agents) and hit the 5-hour quota limit in roughly 10 prompts. This rules out any shared/concurrent usage as a cause — the quota is simply being consumed far too quickly.
All three subscriptions exhibit the same behavior since today. My workflow has not changed.
What Should Happen?
Expected Behavior
On a Max x20 plan, the 5-hour quota window should support a significantly higher number of prompts (historically hundreds of messages). Hitting the limit in ~10 prompts suggests either a change in quota accounting or a backend bug.
Error Messages/Logs
Steps to Reproduce
Log in with a Max x20 subscription on Claude Code (local terminal only)
Send ~10 standard development prompts
Observe that the 5-hour quota is already exhausted
Claude Model
None
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
2.1.76
Platform
Anthropic API
Operating System
Ubuntu/Debian Linux
Terminal/Shell
Terminal.app (macOS)
Additional Information
Related Issues
#28537 — Usage limits hit significantly faster with unchanged workflow
#29178 — Usage limits hit faster since v2.1.53–v2.1.59 (system prompt bloat)
#20767 — Pro quota significantly reduced
#22435 — Inconsistent quota accounting on Max plan (10x burn rate variance documented with mitmproxy)
This issue has 11 comments on GitHub. Read the full discussion on GitHub ↗