Remixer Dec

colibri. Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

UnleashedRecomp-Android. The port was created using Anthropic's Fable 5 AI model. It works EXCLUSIVELY on Adreno GPUs. A port for devices with Mali family GPUs is not planned. I don't understand the technical side myself; I just guided Fable 5 in the right direction, mixed the models, and performed the debugging. Artifacts, freezing, and audio issues are possible.

neuroedge. Run quantized LLMs (GGUF) on ESP32-S3 over UART

Gefen. Gefen: Optimized Stochastic Optimizer

coreai-models. Model export recipes, Python primitives, and Swift runtime utilities for on-device AI

CutterMCP-plus. MCP Server for Cutter

epy-paper-host. A web-based file manager and single-page application (SPA) server packed in a tiny Waveshare ESP32-S3 1.54-inch e-Paper AIoT development board. It's a portable, ultra-low-power, E-Paper Pocket Computer.

Esp32S3. Course and practical examples for learning ESP32-S3 development, including Wi-Fi, Bluetooth, peripherals, and real-world projects.

paroquant. [ICLR 2026] ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference

juggle. CLI Ralph Loops with Good UX

opencode-subagent-output. This plugin pipes output from subagents to the console when "opencode run..." is called and the primary agent invokes a subagent.

paroquant. [ICLR 2026] ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference

llama.cpp-1-bit-turbo. HIP/ROCm fork optimized for AMD RDNA2 (gfx1030) with PrismML Q1_0_G128 1-bit quant support, RotorQuant, TurboQuant, EAGLE3 and P-EAGLE speculative decoding, and full Wave32 kernel optimizations.

C-Asterisk-Alpha. C* (C-Asterisk) is a custom, high-performance programming language. It uses LLVM and native I/O bypasses to run Deep Learning models from scratch 100x faster than Python.

pi-android-re. Android RE skills for Pi coding agent - static (decompile/API extraction) + dynamic (Frida hooking/bypass)

maplibre-gl-js. MapLibre GL JS - Interactive vector tile maps in the browser

openfreemap. Free and open-source map hosting solution with custom styles for websites and apps, using OpenStreetMap data

phosphene. Local generative video panel for Apple Silicon. Wraps LTX-2 MLX, joint audio+video, one-click Pinokio install.

LSFG-Android. Lossless Scaling Frames Generator (LSFG) Application for Android

Rose. 🌹 Rose: Range-Of-Slice Equilibration PyTorch optimizer. Stateless optimization through range-normalized gradient updates.

prodigy. The Prodigy optimizer and its variants for training neural networks.

llama.cpp-tq3. llama.cpp fork with TQ3_1S/4S CUDA kernels — 3.5-bit WHT quantization achieving Q4s quality at 10% smaller size. Based on RaBitQ-inspired Walsh-Hadamard transform. Enables 27B models on 16GB GPUs with 15 tok/s TG, 221 tok/s PP.

llama-cpp-turboquant. LLM inference in C/C++

curl_cffi. Python binding for curl-impersonate fork via cffi. A http client that can impersonate browser tls/ja3/http2 fingerprints.

opencode. The open source coding agent.

codewalk. A native (really fast!!) cross-platform client for OpenCode

vaydns. Go

replsh. The thinking medium for LLM agents

csm.rs. Blazing-fast rust implementation of Sesame's Conversational Speech Model (CSM)

fil-c. Fil-C: completely compatible memory safety for C and C++

coqui-ai-TTS. 🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production

axml2xml. A library for decoding/convert .axml to .xml in JavaScript.

Cataclysm-AOL. Cataclysm: Arsenic and Old Lace — LLM-driven NPCs, far too happy being Teddy Roosevelt.

python-miio. Python library & console tool for controlling Xiaomi smart appliances

AutoAWQ-Reborn. AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:

SlipNet. Android VPN client with DNS tunneling (DNSTT, NoizDNS & Slipstream), NaiveProxy, SSH, Tor, and DoH support — featuring a built-in DNS scanner.

Minecraft-PlayStationPortable-Edition. C++

stretch_audio. Pretty decent algorithm to stretch audio with far less artifacts than WSOLA/librosa.

Awesome-Text-to-3D. A growing curation of Text-to-3D, Diffusion-to-3D works.

inciteful-web. The frontend for Inciteful.xyz.

MedGem. MedGEM brings Google's MedGemma and MedASR to offline Android devices for private, multimodal clinical support anywhere.

littletodo. A tiny menu bar todo app for macOS. Single Swift file, no dependencies.

Plash. 💦 Make any website your Mac desktop wallpaper

claude-usage. macOS menu bar app showing Claude.ai subscription usage limits

rivet. The open-source visual AI programming environment and TypeScript library

librediffusion. C++ / CUDA implementation of StreamDiffusion

macfuse. macFUSE umbrella repository

anet. Simple Rust VPN Client / Server

TheEscapists-j2me. The Escapists fan remake for J2ME

elks. Embeddable Linux Kernel Subset - Linux for 8086

lrzip. Long Range Zip

TimeCapsuleLLM. A LLM trained only on data from certain time periods to reduce modern bias

tts-studio. Test and compare browser-based TTS models!

lpunpack. command-line tool for extracting partition images from super

axml. Android binary XML parser in Python

numAi. A simple AI app for Android 1.0+ with support for deep thinking, image perception, and multiple models. ChatGPT, DeepSeek, Gemini and more for legacy Android

Xml2Axml. This tool can be used to convert binary xml file into human readable xml file and to convert normal xml file into binary xml file.

android-xml-converter. Standalone Implementation of abx2xml and xml2abx

safeshell. Safe shell operations with automatic checkpoints for AI agents. Let agents run freely. Everything is reversible.

vibium. The verification layer for coding agents

apkstudio. Open-source, cross platform Qt6 based IDE for reverse-engineering Android application packages. It features a friendly IDE-like layout including code editor with syntax highlighting support for *.smali code files.

syntax-highlight-element. 👓 Syntax Highlighting using the CSS Custom Highlight API

light-icons. Handpicked collection of premium & light-wighted icons as font

model-explorer. A modern model graph visualizer and debugger

Llama-macOS. A cosy home for your LLMs.

superhtml. HTML Validator, Formatter, LSP, and Templating Language Library

TinyRecursiveModels. Python

DC-Gen. DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space

mbase_nlquery. On-premise, high-performance Text-to-SQL engine for PostgreSQL with single REST API endpoint

APP-android_arm64. Python

ByeByeDPI. Приложение локально запускает ByeDPI и перенаправляет весь трафик через него

hydra. A Nintendo Switch emulator for macOS

rpcsx-ui-android. Kotlin

utcp-specification. The specification for the Universal Tool Calling Protocol

chatllm.cpp. Pure C++ implementation of several models for real-time chatting on your computer (CPU & GPU)

chili3d. A browser-based 3D CAD application for online model design and editing

Westonpack.

vits.cpp. a cpp ggml port of "VITS: Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech." for use in mobile devices. this is my undergraduate project

PySquashfsImage. Python library to read Squashfs image files.

DistillT5. (CVPR 2025) Scailing Down Text Encoders of Text-to-Image Diffusion Models

ik_llama.cpp. llama.cpp fork with additional SOTA quants and improved performance

distroless. 🥑 Language focused docker images, minus the operating system.

psp-ghidra-scripts. Scripts for use with the Ghidra Reverse Engineering Suite

ghidra-allegrex. Ghidra processor module adding support for the Allegrex CPU (PSP)

chaiNNer. A node-based image processing GUI aimed at making chaining image processing tasks easy and customizable. Born as an AI upscaling application, chaiNNer has grown into an extremely flexible and powerful programmatic image processing application.

Kim2091-Models. Models I've trained over the years

llama-cpp-python-cuBLAS-wheels. Wheels for llama-cpp-python compiled with cuBLAS support

firefox-container-proxy. [Firefox extension] Assign a proxy to a Firefox container

tim2. PlayStation 2 TIM2 & CLUT2 format specs.

Rainbow. A texture format converter for different consoles' graphics formats.

glance. A self-hosted dashboard that puts all your feeds in one place

csm. A Conversational Speech Generation Model

gitingest. Replace 'hub' with 'ingest' in any GitHub URL to get a prompt-friendly extract of a codebase

pdd. Small PDD-learning app

CnC_Red_Alert. Command and Conquer: Red Alert

atomize.zed.

apkeep. A command-line tool for downloading APK files from various sources

AntiSplit-M. App to AntiSplit (merge) split APKs (APKS/XAPK/APKM) to regular .APK file on Android

wie. WIPI/SKVM/J2ME Emulator

RustJava. Rust

abliteration. Make abliterated models with transformers, easy and fast

source-map-visualization. A simple visualization of source map data

saba. OpenGL Viewer (OBJ PMD PMX)

py-lmdb. Universal Python binding for the LMDB 'Lightning' Database

munum. Micro 3D Math Library for Rust, JavaScript and WebAssembly

quartz. 🌱 a fast, batteries-included static-site generator that transforms Markdown content into fully functional websites

tiny-rdm. Tiny RDM (Tiny Redis Desktop Manager) - A modern, colorful, super lightweight Redis GUI client for Mac, Windows, and Linux. It also provides a web version that can be deployed via Docker.

noise-shader. An additive cyclic noise shader that makes a cool effect.

elements. Build beautiful, interactive API Docs with embeddable React or Web Components, powered by OpenAPI and Markdown.

simpler-whisper. Dead simple Python bindings for whisper.cpp, all OSs and acceleration supported

AssetStudio. AssetStudioMod - modified version of Perfare's AssetStudio, mainly focused on UI optimization and some functionality enhancements.

HuggingFaceModelDownloader. Simple go utility to download HuggingFace Models and Datasets

zed. Code at the speed of thought – Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.

lew-ui. A beautiful component library based on vue3

mlx-sd-single-file-models. Single safetensors file support Apple MLX Stable Diffusion

stinfo. Safetensors Info command line tool

taskiq. Distributed task queue with full async support

antislop-sampler. Python

Keen-UI. A lightweight Vue.js UI library with a simple API, inspired by Google's Material Design.

python-on-whales. An awesome Python wrapper for an awesome Docker CLI!

photos. Securely encrypt, store, and share files, including but not limited to pictures and videos.

flux-4bit. Jupyter Notebook

dysk. A linux utility to get information on filesystems, like df but better

llama-box. LM inference server implementation based on *.cpp.

rivet. The open-source visual AI programming environment and TypeScript library

webvm. Virtual Machine for the Web

serverless-invoices. Create and manage invoices in your browser. Serverless Invoices is a free invoicing tool for freelancers and small businesses. Made with Vue.js.

matmulfreellm. Implementation for MatMul-free LM.

freej2me. A free cross-platform desktop J2ME emulator with 3D support from J2ME-Loader

llama.sh. No-messing-around sh client for llama.cpp's server

ripgrep. ripgrep recursively searches directories for a regex pattern while respecting your gitignore

httptoolkit. HTTP Toolkit is a beautiful & open-source tool for debugging, testing and building with HTTP(S) on Windows, Linux & Mac :tada: Open an issue here to give feedback or ask for help.

fq. jq for binary formats - tool, language and decoders for working with binary and text formats

android-build. This NDK can run on Linux / MacOS / Windows / FreeBSD / OpenBSD 7.3 / NetBSD for both the x86_64 and AARCH64 architectures.

mdl. Download files and folders recursive from MediaFire

sglang. SGLang is a high-performance serving framework for large language models and multimodal models.

GPT-Over-API. An alternative lightweight ChatGPT client that uses a cost-effective API and eliminates the risk of API token leak by running from an html file in your browser and having no third-party dependencies.

USearch. Fast Open-Source Search & Clustering engine × for Vectors & Arbitrary Objects × in C++, C, Python, JavaScript, Rust, Java, Objective-C, Swift, C#, GoLang, and Wolfram 🔍

distributed-llama. Distributed LLM inference. Connect home devices into a powerful cluster to accelerate LLM inference. More devices means faster inference.

kando. 🌸 Do things with utmost efficiency.

rotaryGames. A set of small games for the m5stickC

AMT. Official code for "AMT: All-Pairs Multi-Field Transforms for Efficient Frame Interpolation" (CVPR2023)

NewPipe-Legacy-Revo. NewPipe with support for older devices

TinyLlama. The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.

llm_poli_compass. A program to automate testing open source LLMs for their political compass scores

ComfyUI-Marigold. Marigold depth estimation in ComfyUI

ComfyUI. The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

llama4micro. A "large" language model running on a microcontroller

vits.cpp. a cpp ggml port of "VITS: Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech." for use in mobile devices. this is my undergraduate project

mm. JavaScript

lm-format-enforcer. Enforce the output format (JSON Schema, Regex etc) of a language model

memfs. Node.js "fs" and browser "File System API" adapters and in-memory implementations 👍

ALMA. State-of-the-art LLM-based translation models.

DALM. Domain Adapted Language Modeling Toolkit - E2E RAG

formvuelar. Vue form components with server-side validation in mind

sd-webui-animatediff. AnimateDiff for AUTOMATIC1111 Stable Diffusion WebUI

tiptap. The headless rich text editor framework for web artisans.

llama2.py. Inference Llama 2 in one file of pure Python

termux-box. Shell

pywebview. Build GUI for your Python program with JavaScript, HTML, and CSS

rumps. Ridiculously Uncomplicated macOS Python Statusbar apps

LLM-Pruner. [NeurIPS 2023] LLM-Pruner: On the Structural Pruning of Large Language Models. Support Llama-3/3.1, Llama-2, LLaMA, BLOOM, Vicuna, Baichuan, TinyLlama, etc.

mlc-chatbot. python interface for mlc chat cli

DPE. [CVPR 2023] DPE: Disentanglement of Pose and Expression for General Video Portrait Editing

SnapCameraPreservation. JavaScript

tinygrad. You like pytorch? You like micrograd? You love tinygrad! ❤️

sd-webui-model-converter. model convert extension for stable-diffusion-webui. supports convert fp16/bf16 no-ema/ema-only safetensors

kohya-trainer. Adapted from https://note.com/kohya_ss/n/nbf7ce8d80f29 for easier cloning

baj-tts. Forsen

GitHubDragonFly.github.io. Number Converter, 3D viewers: 3DM, 3DS, 3D Tiles, 3MF, AMF, ASSIMP, BIM, BREP, COPC, DAE, DST, EPT, FBX, GCODE, GLTF, IFC, IGES, JSON, LAS, LDRAW, MMD, OBJ, PCD, PDB, PLY, PRWM, SPLAT, STEP, STL, SVG, USDZ, VRML, VTK, XYZ and Texture Viewer for PCX, PNG, JPG, BMP, GIF, TIFF, TGA, SVG, WEBP, DDS, KTX(2), EXR, HDR, BASIS, Lottie JSON. Some AR / VR.

gpt-2-3-tokenizer. GPT-2/3 tokenizer based on @latitudegames/GPT-3-Encoder that works in the browser and Deno

dayjs. ⏰ Day.js 2kB immutable date-time library alternative to Moment.js with the same modern API

socket. JavaScript

vite-bundle-visualizer. CLI for visualizing your vite project bundle.

blink. tiniest x86-64-linux emulator

mvp. MVP.css — Minimalist classless CSS stylesheet for HTML elements

hfm. Hugging Face Download (Cache) Manager

InvokeAI. Invoke is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, and serves as the foundation for multiple commercial products.

UnityExplorer. An in-game UI for exploring, debugging and modifying IL2CPP and Mono Unity games.

Hua. Hua is an AI image editor with Stable Diffusion (and more).

nocodb. 🔥 🔥 🔥 A Free & Self-hostable Airtable Alternative

vue-flow. A highly customizable Flowchart component for Vue 3. Features seamless zoom & pan 🔎, additional components like a Minimap 🗺 and utilities to interact with state and graph.

asymmetric-tiling-sd-webui. Asymmetric Tiling for stable-diffusion-webui

stable-diffusion-webui. Stable Diffusion web UI

sygil-webui. Stable Diffusion web UI

Thin-Plate-Spline-Motion-Model. [CVPR 2022] Thin-Plate Spline Motion Model for Image Animation.

stable-diffusion. A latent text-to-image diffusion model

tipsy. tiny playstationy renderer

wails. Create beautiful applications using Go

wepsim. 💡 WepSIM: Web Elemental Processor SIMulator 📱 💻 🖥️

realcugan-ncnn-webassembly. Super-resolution processing for anime images based on WASM; 基于WASM的动漫图片超分辨率处理

gooreplacer. ⚡️⚡️A browser extension to modify HTTP requests :-)

voicevox. 無料で使える中品質なテキスト読み上げソフトウェア、VOICEVOXのエディター

simdjson. Parsing gigabytes of JSON per second : used by Facebook/Meta Velox, the Node.js runtime, ClickHouse, WatermelonDB, Apache Doris, Milvus, StarRocks

kaomojify. 👀 Convert Javascript code to Japanese kawaii kaomoji(かわいい顔文字).

PowerTunnel-Android. Powerful and extensible proxy server with anti-censorship functionality for Android

ordered-json. JavaScript

csv-batch. Streaming CSV parser with no dependencies and has a batch event for lower memory processing in batches as well as a reducer for doing aggregations.

stylegan-xl. [SIGGRAPH'22] StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets

lazydocker. The lazier way to manage everything docker