4 self-hostable options with public code.
Run large language models on your own machine
C/C++ inference for LLMs on ordinary hardware
Drop-in local replacement for the OpenAI API
High-throughput LLM serving engine
