Stars
DeepSeek Harness: Everything is a Plugin.
Waku Waku! Waku Agent is a local-first AI agent harness you actually own, including loop, memory, eval, all in code built to stay legible as it grows.
🐧 Harness for RSI. Let AI Build AI
An agentic skills framework & software development methodology that works.
分享AI Infra知识&代码练习：PyTorch、vLLM/SGLang、slime/vime框架入门⚡️、性能加速🚀、大模型基础🧠、AI软硬件🔧等
GPT Image 2 prompt gallery, image prompt library, agentic skill, and CLI for OpenAI image generation/editing
Code and Data for paper "GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?"
A tutorial on modern GPU programming for machine learning systems
A high-performance RL training-inference weight synchronization framework, designed to enable second-level parameter updates from training to inference in RL workflows
This code implements the algorithm of FIPO, a value-free RL recipe for eliciting deeper reasoning from a clean base model.
Skills for research Project and AI model degsin
GlobalDentBench: A Multinational Benchmark for Evaluating LLM Clinical Reasoning in Dentistry with Expert Calibration
Agentifying Patient Dynamics within LLMs through Interacting with Clinical World Model
A curated list of resources (surveys, papers, benchmarks, and opensource projects) on Rubrics
[pip install medmnist] 18x Standardized Datasets for 2D and 3D Biomedical Image Classification
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…
Hundreds of agent skills for medical research, including protocol design, data analysis, evidence insights, and academic writing.
MedEvalKit: A Unified Medical Evaluation Framework
slime is an LLM post-training framework for RL Scaling.
Mobile and Web client for Codex and Claude Code, with realtime voice, encryption and fully featured
An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale
Flash Attention from Scratch on CUDA Ampere
A bag of training monitor skills for model training.
Data processing for and with foundation models! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷
Minimalistic large language model 3D-parallelism training
A research intelligence agent pipeline for daily paper and blog triage to your email inbox.
Nano vLLM with vLLM v1's request scheduling strategy and chunked prefill