Stars
The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!
AutoScientist: internal architecture & API reference (compiled from observed behavior)
Browser-based ground station suite for satellite tracking, SDR reception, hardware control, and telemetry decoding
Terminal protein structure viewer — interactive 3D visualization of PDB/mmCIF structures with cartoon ribbons, braille rendering, and Sixel/Kitty graphics
A lightweight inference engine supporting speculative speculative decoding (SSD).
SkillsBench evaluates how well skills work and how effective agents are at using them.
AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
Scripts for exporting ONNX models from the Pocket TTS PyTorch library and weights.
A community database for the problems on the erdosproblems.com site
Soprano: Instant, Ultra-Realistic Text-to-Speech
Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).
Zellij plugin that exports pane names to JSON for shell integration
An extremely fast Python type checker and language server, written in Rust.
Fully automatic censorship removal for language models
Modified version of Chatterbox that accepts text files as input and no character restrictions. I use it to make audiobooks, especially for my kids.
An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System
A lightweight Python package for Automatic Speech Recognition using ONNX models
invisiofficial / rk-llama.cpp
Forked from ggml-org/llama.cppLlama.cpp with the Rockchip NPU integration as a GGML backend.
Rust multi‑backend OCR/VLM engine (DeepSeek‑OCR-1/2, PaddleOCR‑VL, DotsOCR) with DSQ quantization and an OpenAI‑compatible server & CLI – run locally without Python.
A modular graph-based Retrieval-Augmented Generation (RAG) system