Prompt caching
4 mentions across all digests
Claude API optimization technique that reuses cached context across requests, with suspected bugs in Claude Code causing 10-20x token cost inflation for users exhausting usage limits faster than expected.
Claude Code users hitting usage limits 'way faster than expected'
Claude Code Pro subscribers are burning through monthly quotas in just 12 days, with suspected prompt caching bugs inflating token costs by 10-20x.
Anthropic admits Claude Code users hitting usage limits 'way faster than expected'
Claude Code users are exhausting monthly limits 10-20x faster than expected due to prompt caching bugs, leaving Pro subscribers locked out after just 12 of 30 days.
Claude Code cache confusion as Anthropic tweaks defaults, but quotas still drain
Anthropic's rollback of Claude Code's prompt cache TTL from one hour to five minutes contradicts internal cost assurances, as users report measurably faster quota depletion from reduced cache reuse.
Pro Max 5x Quota Exhausted in 1.5 Hours Despite Moderate Usage
Anthropic's Claude API charges cached prompt tokens at full rate instead of the promised 1/10 discount, causing Pro Max users to exhaust quotas within hours rather than benefiting from semantic caching cost savings.