Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
cache-coolgithub.comCache-Cool is a simple LLM caching proxy that sits in front of LLM API calls to save redundant requests, improving performance and reducing costs. It supports multiple LLM providers (e.g., OpenAI, Claude, Groq) and offers dynamic configuration via a /configure endpoint. Cached responses are stored using MongoDB and JSON, with features like LRU eviction to manage cache size.
Parse Score
