从 65,707 个高星开源项目里，
找到你要的下一个。

✍🏻 Source Code Deep Dives, System Design & Engineering Blogs | Halfrost-Field 冰霜之地：源码解析、系统设计与工程实践笔记

✍🏻 Source Code Deep Dives, System Design & Engineering Blogs | Halfrost-Field 冰霜之地：源码解析、系统设计与工程实践笔记

kimi-k3-in-c 是用 C 编写的 AI 工具，用于大模型与智能体应用开发。

A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.

FedML 是用 Python 编写的学习资料，用于系统学习与查阅，主打可伸缩、支持自托管。

FEDML - The unified and scalable ML library for large-scale distributed training, model serving, and federated learning. FEDML Launch, a cross-cloud…

校招、秋招、春招、实习好项目！带你从零实现一个高性能的深度学习推理库，支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning i…

校招、秋招、春招、实习好项目！带你从零实现一个高性能的深度学习推理库，支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning inference library step by step

MTPLX 是用 Python 编写的 AI 工具，用于大模型与智能体应用开发，主打速度快。

The fastest way to run Qwen 3.8 Flash Next, Qwen 3.8 27B and Ternary Bonsai 2 27B on a Mac: 125 tok/s in OpenCode on an M5 Max, and a 27B model on 1…

grule-rule-engine 是用 Go 编写的 AI 工具，用于大模型与智能体应用开发。

Rule engine implementation in Golang

onediff 是用 Jupyter Notebook 编写的 AI 工具，用于大模型与智能体应用开发。

OneDiff: An out-of-the-box acceleration library for diffusion models.

sonar 是用 Python 编写的 AI 工具，用于人工智能方向的建模与推理。

Large-scale LLM inference engine

kvcached 是用 Python 编写的 AI 工具，用于大模型与智能体应用开发。

Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond

xllm 是用 C++ 编写的 AI 工具，用于大模型与智能体应用开发，主打高性能。

A high-performance inference engine for LLM, VLM, DiT and REC models, optimized for diverse AI accelerators. It is hosted in OpenAtom Foundation.

nobodywho 是用 Rust 编写的游戏开发项目，用于游戏与引擎开发。

NobodyWho is an inference engine that lets you run LLMs locally and efficiently on any device.

zero-to-sglang 是用 Python 编写的 AI 工具，用于大模型与智能体应用开发。

Official SGLang x Datawhale course on LLM inference: understand inference, build a mini-sglang from scratch, then read the real SGLang source and la…

FeatherCNN 是用 C++ 编写的移动端方案，用于移动应用开发，主打高性能。

FeatherCNN is a high performance inference engine for convolutional neural networks.

ai-hub-models 是用 Python 编写的学习资料，用于系统学习与查阅。

Qualcomm® AI Hub Models is our collection of state-of-the-art machine learning models optimized for performance (latency, memory etc.) and ready to …

Paddle.js 是用 JavaScript 编写的学习资料，用于系统学习与查阅，主打开源免费。

Paddle.js is a web project for Baidu PaddlePaddle, which is an open source deep learning framework running in the browser. Paddle.js can either load…