为微调大模型生成高质量合成数据的工具。
AI 开源项目
精选 GitHub 上值得关注的 AI 开源项目:大模型、智能体、RAG、本地部署、图像与语音……共收录 1038 个。
想找可以部署在 Cloudflare 免费套餐上的项目?看 Cloudflare 开源
Reference implementations of MLPerf® inference benchmarks
A fast and lightweight framework for creating decentralized agents with ease.
The backend agents build with - Multimodal database, orchestration, and serving in one file
Fast Multimodal LLM on Mobile Devices
A high-performance inference engine for LLM, VLM, DiT and REC models, optimized for diverse AI accelerators. It is hosted in OpenAtom Foundation.
Chat with your data - AI data analysis and visualization on CSV, Postgres, MySQL, Snowflake, SQLite...
Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.
A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization and pruning.
AIDE: an LLM agent for machine learning engineering - the research Weco grew out of. Referenced in OpenAI MLE-bench.
KAI Scheduler is an open source Kubernetes Native scheduler for AI workloads at large scale
A Data Streaming Library for Efficient Neural Network Training
NobodyWho is an inference engine that lets you run LLMs locally and efficiently on any device.
基于本地 llama.cpp 的 VS Code 补全与对话插件。
The LLM Anti-Framework
Self-hosted LLM router with a managed safety net. OpenAI-compatible. BYOK. Single-workspace. Streaming. For more advanced routing choose hosted OrcaRouter
MemFree - Hybrid AI Search Engine & AI Page Generator
The open document intelligence platform for builders and hackers - DMS for the agentic world
AuK: An Open-Source Foundational Model for Speech Generation and Editing
Ouroboros — self-creating AI agent. Born Feb 16, 2026.
KaibanJS is a JavaScript-native framework for building and managing multi-agent systems with a Kanban-inspired approach.
Platform for AI-powered software engineers
AI 小说创作软件:把灵感、角色、世界观、大纲、章节写作、审稿和修稿组织成可控流程;提供 Windows/macOS 桌面版,支持本地和在线模型。AI Novel Writing Software: Organizes inspirations, characters, worldbuilding, outlines, chapter drafting, review, and revision into a controllable
A benchmark built to evaluate and improve agent capabilities for supporting legal work.