May 1
Add Anthropic prompt caching (cache_control) for Claude API calls
Hey team, Big fan of v2.13's BYOK Claude integration! Quick request that should be a high-impact, low-effort win: **Problem** My translation prompt is 618 lines / 35KB (custom rules for slang, batch codes, customer rapport, etc.). Every time I trigger SuperWhisper, the full prompt is sent to Claude. Time-to-first-token is noticeably slow on Sonnet 4.6 because of this. **Ask** Add `cache_control: { type: "ephemeral" }` to the system prompt block in your Anthropic API call. **Impact** - ~90% reduction in input token costs (huge savings on your end since you're the API biller for subscription users) - 2–3x faster TTFT on cache hits (5-min TTL matches typical dictation cadence perfectly) - Single-line code change in your backend - Anthropic officially recommends this for any system prompt > 1KB **Docs:** https://docs.claude.com/en/docs/build-with-claude/prompt-cac… Happy to beta-test on my account. Thanks for building such a great app!
Pending
