Cutting LLM inference costs by 36% with prompt caching

1 points | by lizakatz an hour ago

No comments yet