-
Tsinghua University
- Beijing, China
-
02:48
(UTC +08:00)
Highlights
- Pro
Stars
The absolute trainer to light up AI agents.
A selective knowledge distillation algorithm for efficient speculative decoders
A scalable asynchronous reinforcement learning implementation with in-flight weight updates.
wyf9661 / typora-free
Forked from zogodo/typora-0.11.18typora-0.11.18 (last free version)
EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL
一个让 AI 模型在真实市场中进行实盘交易与对抗的实验平台。目标是通过不断迭代,让智能体真正学会在不确定市场中生存和盈利。
SkyRL: A Modular Full-stack RL Library for LLMs
Bridge Megatron-Core to Hugging Face/Reinforcement Learning
Allow torch tensor memory to be released and resumed later
Post-training with Tinker
[EMNLP 2025 Demo] PDF scientific paper translation with preserved formats - 基于 AI 完整保留排版的 PDF 文档全文双语翻译,支持 Google/DeepL/Ollama/OpenAI 等服务,提供 CLI/GUI/MCP/Docker/Zotero
chat log tool, easily use your own chat data. 聊天记录工具,轻松使用自己的聊天数据
All-in-One Sandbox for AI Agents that combines Browser, Shell, File, MCP and VSCode Server in a single Docker container.
General technology for enabling AI capabilities w/ LLMs and MLLMs
Excalidraw-CN 是支持中文手写和多画布的 Excalidraw 白板工具。Excalidraw-CN is a whiteboard supporting Chinese hand draw font and multi-canvas based on Excalidraw.
Generate interactive call graphs for various languages
Code and weights for the paper "Cluster and Predict Latents Patches for Improved Masked Image Modeling"
Minimalistic large language model 3D-parallelism training
Fast, Flexible and Portable Structured Generation
Implementation of a methodology that allows all sorts of user defined GPU kernel fusion, for non CUDA programmers.
Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (LLM).
Towards a Unified View of Large Language Model Post-Training
Checkpoint-engine is a simple middleware to update model weights in LLM inference engines
tile-ai / tilescale
Forked from tile-ai/tilelangTile-based language built for AI computation across all scales