Starred repositories
Accelerating MoE with IO and Tile-aware Optimizations
Machine Learning Engineering Open Book
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.
An Open Phone Agent Model & Framework. Unlocking the AI Phone for Everyone
微舆：人人可用的多Agent舆情分析助手，打破信息茧房，还原舆情原貌，预测未来走向，辅助决策！从0实现，不依赖任何框架。
Building a inclusive, scalable, and high-performance multilingual translation model
Language Savant. If your repository's language is being reported incorrectly, send us a pull request!
Collect, organize, use, and share, all in OmniBox.
slime is an LLM post-training framework for RL Scaling.
🥢像老乡鸡🐔那样做饭。主要部分于2024年完工，非老乡鸡官方仓库。文字来自《老乡鸡菜品溯源报告》，并做归纳、编辑与整理。CookLikeHOC.
A Next-Generation Training Engine Built for Ultra-Large MoE Models
Awesome MCP Servers - A curated list of Model Context Protocol servers
Paper list for Efficient Reasoning.
Ring is a reasoning MoE LLM provided and open-sourced by InclusionAI, derived from Ling.
[COLING 2025] Let's Focus on Neuron: Neuron-Level Supervised Fine-tuning for Large Language Model
Ongoing research training transformer models at scale
An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.
A version of verl to support diverse tool use
My learning notes for ML SYS.
Kimi K2 is the large language model series developed by Moonshot AI team
Code for "Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate" [COLM 2025]
Unified the Anonymous and Camera Ready Version, hope everyone can get an ACCEPT
学术双语简历模板，涵盖教育背景、论文发表、项目经历、竞赛经历和个人陈述等关键部分，可适用于申请研究生项目、学术职位或相关行业岗位。
RewardAnything: Generalizable Principle-Following Reward Models