semantic-cache

https://github.com/stokry/semantic-cache

Cache LLM responses using semantic similarity matching. Similar questions return cached answers instantly, cutting API costs by 70% or more. Works with OpenAI, Anthropic, and Gemini.