"Usage credits are required for long context requests" fires on Claude Pro despite low usage (43% session / 15% weekly)

Status Open
Reported on v2.1.126
Maintainer reply None cached
Activity 0 comments · opened Aug 13, 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?

Preflight Checklist

  • [x] I have searched existing issues and this appears related to #61808, #62314, #62143 but is a fresh reproduction with usage data attached
  • [x] This is a single bug report
  • [x] I am using the latest version of Claude Code

What's Wrong?

Every request fails with the error Usage credits are required for long context requests. — this happens repeatedly, including on /continue and /compact, even though account usage is nowhere near any plan limit.

Environment

  • Plan: Claude Pro
  • Auth method: Claude AI (OAuth)
  • Platform/OS: [fill in — e.g. Windows/macOS/Linux]
  • Claude Code version: [fill in — output of claude --version]
  • Terminal: [fill in]

Usage at time of error (from Account & Usage panel)

  • Session (5hr): 43% used, resets in 3h
  • Weekly (7 day): 15% used, resets in 4d

Steps to Reproduce

  1. Start a normal Claude Code session on a Pro plan account
  2. Send a prompt (context usage low, well under plan limits)
  3. Observe error: Usage credits are required for long context requests.
  4. Try /continue — same error
  5. Try /compact — same error

Expected Behavior

Standard-context requests on a Pro plan account with usage well below both session (43%) and weekly (15%) limits should not require prepaid usage credits.

Actual Behavior

Every request is blocked by the long-context credit gate regardless of actual token usage, making Claude Code unusable despite ample remaining plan allowance.

Additional Context

This matches the pattern reported in #61808, #62314, and #62143 — the long-context billing tier appears to be triggered incorrectly (false positive) rather than reflecting genuine context/token overage. Screenshots attached showing (1) the repeated error message and (2) the Account & Usage panel confirming usage is low.

What Should Happen?

What's Wrong?

Every request fails with the error Usage credits are required for long context requests. — this happens repeatedly, including on /continue and /compact, even though account usage is nowhere near any plan limit.

Environment

  • Plan: Claude Pro
  • Auth method: Claude AI (OAuth)
  • Platform/OS: windows
  • Claude Code version: 2.1.126
  • Terminal: [fill in]

Usage at time of error (from Account & Usage panel)

  • Session (5hr): 43% used, resets in 3h
  • Weekly (7 day): 15% used, resets in 4d

Steps to Reproduce

  1. Start a normal Claude Code session on a Pro plan account
  2. Send a prompt (context usage low, well under plan limits)
  3. Observe error: Usage credits are required for long context requests.
  4. Try /continue — same error
  5. Try /compact — same error

Expected Behavior

Standard-context requests on a Pro plan account with usage well below both session (43%) and weekly (15%) limits should not require prepaid usage credits.

Actual Behavior

Every request is blocked by the long-context credit gate regardless of actual token usage, making Claude Code unusable despite ample remaining plan allowance.

Additional Context

This matches the pattern reported in #61808, #62314, and #62143 — the long-context billing tier appears to be triggered incorrectly (false positive) rather than reflecting genuine context/token overage. Screenshots attached showing (1) the repeated error message and (2) the Account & Usage panel confirming usage is low.

Error Messages/Logs

Steps to Reproduce

Steps to Reproduce

  1. Start a normal Claude Code session on a Pro plan account
  2. Send a prompt (context usage low, well under plan limits)
  3. Observe error: Usage credits are required for long context requests.
  4. Try /continue — same error
  5. Try /compact — same error

Claude Model

Sonnet (default)

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.126

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

PowerShell

View original on GitHub ↗