Python

A Model Context Protocol (MCP) server implementation that provides database capabilities for Chroma

ComfyUI adaptation of IDM-VTON for virtual try-on.

[NeurIPS 2022] 🛒WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents

Modular Macro Keyboard System

Open-Source Reproduction/Demo of the LLM Riddles Game

三元三小时手敲大模型

🌕 [BMVC 2022] You Only Need 90K Parameters to Adapt Light: A Light Weight Transformer for Image Enhancement and Exposure Correction. SOTA for low light enhancement, 0.004 seconds try this for pre-processing.

Flux.2Klein 9B Enhancement Nodes Suite

Turns a Raspberry Pi into a mobile backup and media management hub. Backs up files from cameras and/or storage devices to local drives and/or the cloud, aggregates multiple sources and can forward them for processing. View, rate, and selectively share your media on the go, including to social networks.

WARNING: THE RIETVELD SERVER MAY DISAPPEAR! See top of README.md. Code Review, hosted on Google App Engine

Blender client for libsm64

Open Data Cube analyses continental scale Earth Observation data through time

猫蛋儿安全团队编写的poc能报就能打。企业微信、海康、Metabase、Openfire、泛微OA......

Agentic RAG for local and self-hosted document search: hybrid retrieval, reranking and multimodal RAG on embedded LanceDB, with Docling parsing and an MCP server

Template for Python Projects

[ICLR2026] Official implementation for "JanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation"

Access your Taskwarrior tasks from any browser anywhere

Cutadapt removes adapter sequences from sequencing reads

Python interface for the Firmata (http://firmata.org/) protocol. It is compliant with Firmata 2.1. Any help with updating to 2.2 is welcome. The Capability Query is implemented, but the Pin State Query feature not yet.

gather gather gather

Driving lessons made simpler. Custom scheduling API built with Python.

A Python wrapper around Synology API

[ECCV2024] Grounded Multimodal Large Language Model with Localized Visual Tokenization

huggingface mirror download

[ICLR 2026] Streamlining Cartoon Production with Generative Post-Keyframing

A lightweight, AI-native training framework for large language models. Designed for fast iteration, reproducible experiments, and modular configuration across SFT, RLVR, and evaluation workflows.

The fastest digital human algorithm, now on your desktop.

Minimal Jupyter C kernel