vLLM

Pinned Loading

1. vllm vllm Public

   A high-throughput and memory-efficient inference and serving engine for LLMs

   Python 89.5k 21k

2. vllm-omni vllm-omni Public

   A framework for efficient model inference with omni-modality models

   Python 6.2k 1.5k

3. recipes recipes Public

   Common recipes to run vLLM

   JavaScript 981 378

4. llm-compressor llm-compressor Public

   Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM

   Python 3.7k 625

5. speculators speculators Public

   A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM

   Python 750 190

6. semantic-router semantic-router Public

   A programmable Mixture-of-Models router for heterogeneous LLM inference

   Go 5.2k 822