Pearlite

Pearlite とは？

Every LLM API call costs money even when you've answered the exact same question 1,000 times before. Pearlite sits between your app and any LLM, intercepts semantically similar queries, and returns cached responses in <12ms instead of burning another API call. Not exact-match caching, semantic caching. "What's the price?" and "how much does it cost?" are the same question. Drop in 2 lines of code, keep your existing provider, and cut LLM spend by up to 60%.