vllm-ascend by vllm-project

High-performance Inference and Deployment Toolkit for LLMs and VLMs based on PaddlePaddle. Maintained by PaddlePaddle on GitHub, where it has earned 3,703 stars from the community.

The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separation, cloud routing. Drop-in OpenAI replacement. Works with Claude Code, Cursor, Aider. Maintained by raullenchai on GitHub, where it has earned 3,400 stars from the community.

A GPU cluster manager for high-performance AI model serving (vLLM, SGLang) and on-demand SSH-accessible GPU instances. Maintained by gpustack on GitHub, where it has earned 5,436 stars from the community.