Pro Max 5x Quota Exhausted in 1.5 Hours Despite Moderate Usage
Summary
This GitHub issue documents quota exhaustion on Claude Code Pro Max 5x after 1.5 hours of moderate usage. It argues that cache_read tokens may count at full rate against quotas, negating the cost benefits of prompt caching, and notes background sessions, auto-compact spikes, and a 1M context window amplifying token usage. It provides reproduction steps, environment details, and suggested improvements for quota visibility and fairness.