ToolAI.io · GitHub 频道
一个基于事实的 LLM、智能体、MCP、RAG 和 AI 开发项目索引；每个条目均包含许可证、设置、下载、截图及相关资源。
公开仓库数据
NVIDIA/Model-Optimizer
A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
NVIDIA/warp
把带类型的 Python 函数编译为计算内核，提供几何、物理仿真和可微计算能力，可接入机器学习流程。
NVIDIA/Megatron-LM
Ongoing research training transformer models at scale
openai/openai-agents-js
A lightweight, powerful framework for multi-agent workflows and voice agents
NVIDIA/aistore
AIStore: scalable storage for AI applications
google-deepmind/mujoco
Google DeepMind 维护的多关节接触动力学引擎，提供模型描述、物理步进、交互查看器和 Python 接口，适合机器人、控制与强化学习环境开发。
hydra-ecosystem/hydra
面向 Python 应用的配置框架，通过配置组合、命令行覆盖和多任务运行管理实验，适合把数据、模型与训练参数整理成可复用的配置结构。
Tencent-Hunyuan/UniRL
UniRL is a Framework for Unified Multimodal Model Reinforcement Learning
PaddlePaddle/PaddleFormers
PaddleFormers is an easy-to-use library of pre-trained large language model zoo based on PaddlePaddle.
xai-org/grok-build
SpaceXAI's coding agent harness and TUI. Fullscreen, mouse interactive, extensible.
PaddlePaddle/Paddle
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）
anthropics/claude-code-action
将 Claude Code 接入 GitHub Actions，支持评论触发、代码审查和按提示执行维护任务，可配置不同模型服务提供方。
第 19 / 26 页 · 305 个项目
