This page provides information about LLooM, a tool that uses raw LLM logits to weave threads in a probabilistic way. It includes instructions on how to use LLooM with various environments, such as vLLM, llama.cpp, and OpenAI. The README also explains the parameters and configurations for LLooM.
