Misleading error message when hitting 32000 token maximum limit

Resolved 💬 3 comments Opened Jan 15, 2026 by joonlab Closed Jan 19, 2026

Description

When Claude's response exceeds the maximum output token limit (32,000), the error message suggests configuring CLAUDE_CODE_MAX_OUTPUT_TOKENS environment variable. However, this suggestion is unhelpful because 32,000 is already the hard limit and cannot be increased.

Current Behavior

Error message:

API Error: Claude's response exceeded the 32000 output token maximum. To configure this behavior, set the CLAUDE_CODE_MAX_OUTPUT_TOKENS environment variable.

This message implies that setting the environment variable could resolve the issue, but:

  • 32,000 is the maximum allowed value
  • Setting it higher (e.g., 64000) results in validation error
  • The suggestion is misleading when already at the limit

Expected Behavior

When the limit is already at maximum (32,000), the error message should provide actionable guidance, such as:

API Error: Claude's response exceeded the 32000 output token maximum. 
This is the maximum allowed limit. Consider breaking your request into smaller parts.

Suggestion

Add conditional logic to the error message:

  • If current limit < 32,000: Suggest increasing CLAUDE_CODE_MAX_OUTPUT_TOKENS
  • If current limit = 32,000: Suggest alternative solutions (e.g., splitting tasks)

Environment

  • Claude Code CLI
  • Observed with default settings (no custom CLAUDE_CODE_MAX_OUTPUT_TOKENS set)

View original on GitHub ↗

This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗