Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-R1, GLM-5, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Llava, Phi4, ...)…

FinGLM 财报问答标注分类

专门面向中文上市公司财报 PDF的预处理工具，用来把复杂的 PDF 文档解析成 结构化的「父子分块 + 表格组」JSON 数据，方便后续做向量检索、问答系统、风控分析等。

Wallpaper engine PKG extractor/TEX to image converter

A large-scale face dataset for face parsing, recognition, generation and editing.

Spriteworld: a flexible, configurable python-based reinforcement learning environment

Burgess et al. "MONet: Unsupervised Scene Decomposition and Representation"

Mamba SSM architecture

An interpreter for RASP as described in the ICML 2021 paper "Thinking Like Transformers"

(CVPR 2025) Switti: Designing Scale-Wise Transformers for Text-to-Image Synthesis

基于本地大模型Ollama的PDF问答对提取工具

Latex code for making neural networks diagrams

Implementation of Slot Attention from GoogleAI

Multi-object image datasets with ground-truth segmentation masks and generative factors.

Flash浏览器 / Flash Browser

A list of all repositories from Music X Lab

A modern Fluent Design replacement for the old Metro themed flyouts present in Windows.

Beautiful icons for your favourite terminal apps like Hyper and iTerm2

Play and synthesize MIDI to audio - easy to use Python/CLI API to FluidSynth.

ElevenClock: Customize Windows 11 taskbar clock

Implementation of Parti, Google's pure attention-based text-to-image neural network, in Pytorch

Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network, in Pytorch

Large Concept Models: Language modeling in a sentence representation space

A much modified lyric editor with fancy (maybe) UI based on WPF

输入法状态管理: 实时提示 + 状态切换

Kandinsky 2 — multilingual text2image latent diffusion model

OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.