Micro-Agent: Beat Frontier Models with Collaboration Inside Model API(vllm.ai)64 points by matt_d 3 months ago | 19 comments
Eagle 3.1: Collaboration Between the EAGLE Team, vLLM Team, and TorchSpec Team(vllm.ai)60 points by berlianta 5 months ago | 20 comments
vLLM: Easy, Fast, and Cheap LLM Serving with PagedAttention(vllm.ai)260 points by wskwon 3 years ago | 39 comments