LangCache
Semantic cache for LLM apps.
Cut LLM latency and token spend by serving cached answers to semantically similar prompts, powered by Redis.
A near-hit on a previous question returns instantly instead of calling the model again. Drop-in in front of any LLM app.