Stars
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
Video OSINT agent: senses + OSINT reach for any agent.
Code and data release for the paper "Seeing the Arrow of Time in Large Multimodal Models"
Official agent skills for coding agents working with the Cloudglue API / SDK
Config-driven hook framework for Claude Code — guards, analytics, coaching, and interactive TUI from one YAML file
A prompt engineering functional programming language
A vibe coded threejs shooter arena using threejs agent skills
A vibe coded 3D game inspired by the Mario Party mini game. Made with Codex CLI and Agent Skills
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Wrapper of 50+ image matching models with a unified interface
The batteries-included agent harness.
Cameradar hacks its way into RTSP videosurveillance cameras
AMY - A high-performance fixed-point Music synthesizer librarY
A 15TB Collection of Physics Simulation Datasets
repo for gen ai music workshop project - balkon
Fun MCP server that draws / generates images and saves the output locally
A desktop application that adds powerful tools to Claude and AI platforms
Official MCP server for Cloudglue API
Official OpenAPI specification for Clouglue API
Official JavaScript / TypeScript SDK for Cloudglue API
YuE: Open Full-song Music Generation Foundation Model, something similar to Suno.ai but open
Tegridy MIDI Dataset for precise and effective Music AI models creation.
Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.