MNBVC(Massive Never-ending BT Vast Chinese corpus)超大规模中文语料集。对标chatGPT训练的40T数据。MNBVC数据集不但包括主流文化，也包括各个小众文化甚至火星文的数据。MNBVC数据集包括新闻、作文、小说、书籍、杂志、论文、台词、帖子、wiki、古诗、歌词、商品介绍、笑话、糗事、聊天记录等一切形式的纯文本中文数据。

Taming Transformers for High-Resolution Image Synthesis

Animation engine for explanatory math videos

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.5, DeepSeek-R1, GLM-5, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Llava, Phi4, ...)…

Nano vLLM

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

XLeRobot: Practical Dual-Arm Mobile Home Robot for $660

Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.

LeetGPU Challenges

Native Multimodal Models are World Learners

Documentation that simply works

Renderer for the harmony response format to be used with gpt-oss

A framework for few-shot evaluation of language models.

Flexible Python configuration system. The last one you will ever need.

Hydra is a framework for elegantly configuring complex applications

Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.

Selftok: Discrete Visual Tokens of Autoregression, by Diffusion, and for Reasoning

verl: Volcano Engine Reinforcement Learning for LLMs

所有小初高、大学PDF教材。

为键盘工作者设计的单词记忆与英语肌肉记忆锻炼软件 / Words learning and English muscle memory training software designed for keyboard workers

Distributed Compiler based on Triton for Parallel Systems

Generation of diagrams like flowcharts or sequence diagrams from text in a similar manner as markdown

Easy-to-use and powerful LLM and SLM library with awesome model zoo.

A fast communication-overlapping library for tensor/expert parallelism on GPUs.

CUDA Library Samples

Main libjpeg-turbo repository

PyExifTool (active PyPI project) - A Python library to communicate with an instance of Phil Harvey's ExifTool command-line application. Runs one process with special -stay_open flag, and pipes data…