Show HN: Do you want a semantic cache for free with zero code changes?

Khazad is an open-source semantic cache for LLM (large language model) API calls — it lets developers reuse responses to semantically similar questions without rewriting any code. - LLM APIs (like OpenAI, Anthropic, etc.) charge per request, and many queries are near-duplicates (e.g., "What is Rust?" vs "Tell me about the Rust programming language"). A semantic cache detects this similarity and returns the cached answer, saving cost and latency. - It works by intercepting API calls via a proxy, so no code changes are needed — a significant practical advantage. - "Semantic" means it uses embeddings (numerical representations of meaning) to judge similarity, not exact string matching. - This is relevant to anyone building AI-powered products, especially as LLM usage scales and costs become a concern.