API rate limits on the Claude platform are raised across the board, and tiers no longer depend on API spend. The latest Sonnet and Haiku models get 5x higher rate limits at the top tier.
Original text +
Claude 平台 API 速率限制整体提高,层级不再按 API 支出划分。最新的 Sonnet 和 Haiku 模型在最高层级拿到 5 倍速率限制。
A change to the limit rules themselves, affecting your allowance going forward.
// CONTEXT
- Most recent Claude Code record: 2026-09-01
- Average gap 12.1d · longest 47.6d · 78% of past waits were shorter than this one