[BUG]

Environment

• Platform (select one):
• [x] Anthropic API
• [ ] AWS Bedrock
• [ ] Google Vertex AI
• [ ] Other: <!-- specify -->
• Claude CLI version: <!-- output of claude --version --> 1.0.15 (Claude Code)
• Operating System: <!-- e.g. macOS 14.3, Windows 11, Ubuntu 22.04 --> macOS 15.5
• Terminal: <!-- e.g. iTerm2, Terminal App --> macOS terminal app

Bug Description

Adding any large, but acceptably large, code file to context results in the following error (note, I removed the org id and replaced with [some-uuid].

API Error (429 {"type":"error","error":{"type":"rate_limit_error","message":"This request would exceed the rate limit for your organization ([some-uuid]) of 20,000 input tokens per minute. For details, refer to: https://docs.anthropic.com/en/api/rate-limits. You can see the response headers for current usage. Please reduce the prompt length or the maximum tokens requested, or try again later. You may also contact sales at https://www.anthropic.com/contact-sales to discuss your options for a rate limit increase."}}) · Retrying in 76 seconds… (attempt 1/10)

I tried asking CC about this directly, and specifically asking if my CLAUDE.md files were too long. There didn't seem to be any 'wrong' way I was using CC, nor were my CLAUDE.md files too long (tho it rearranged them for me and that didn't help at all).

One of the files it's trying to access is about 800 lines, but they are sparse lines with not all that much code actually. A file this size should be totally reasonably accessible to CC.

I'm using a personal API directly with CC, have enough credit, yet because of this error CC is mostly unusable unless I tread extremely carefully.

Steps to Reproduce

1. set up CC with an API key with Tier 1 rate limits
2. ask CC to modify a code file that is not small

Expected Behavior

CC should not have a rate limit error, especially when it's credits I purchased and would simply like to use.

Actual Behavior

The 429 error happens