Lists (13)
Starred repositories
[ICML 2026] The official implementation of "Mean Flow Distillation: Robust and Stable Distillation for Flow Matching Models".
A beautiful, simple, clean, and responsive Jekyll theme for academics
Minima is a one-size-fits-all Jekyll theme for writers.
The source code that powers readthedocs.org
A hands-on course for building modern LLMs from scratch in PyTorch, with 23 runnable Jupyter Notebooks covering tokenizers, attention, MoE, RLHF, inference, evaluation, and distillation.
A 3B-active-parameter native unified multimodal model for image and video understanding, generation, and editing.
Python best practices guidebook, written for humans.
👻 An elegant ghost blogging theme. Mandarin optimization support.
Independent technology for modern publishing, memberships, subscriptions and newsletters.
A framework for efficient model inference with omni-modality models
An implementation of Model Context Protocol (MCP) server for Argo CD.
Implement a reasoning LLM in PyTorch from scratch, step by step
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
slime is an LLM post-training framework for RL Scaling.
Multimodal RL training framework for diffusion & omni models
[NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL
NVFP4 Flash-Attention 4 on BlackWell
AG2 (formerly AutoGen): The Open-Source AgentOS.Join us at: https://discord.gg/sNGSwQME3x
《软件设计的哲学》中文翻译 | Chinese translation of A Philosophy of Software Design
CUDA Templates and Python DSLs for High-Performance Linear Algebra
cuda-oxide is an experimental Rust-to-CUDA compiler that lets you write (SIMT) GPU kernels in safe(ish), idiomatic Rust. It compiles standard Rust code directly to PTX — no DSLs, no foreign languag…