CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies

Official code for MAMMA: Markerless Accurate Multi-person Motion Acquisition.

The repository provides code for running inference with the SAM 3D Body Model (3DB), links for downloading the trained model checkpoints and datasets, and example notebooks that show how to use the…

Import a 3D Model and automatically assign and export animations

ESP-IDF project that build LightMyInk Binaries

Examples and demos for the Vulkan C++ API

Blender Add-on for importing and exporting Shader Nodes with a text

A minimal from scratch arduino library for the c3 programming language written without any external dependencies

Offline MMO private servers on Steam Deck. Free guides, Docker scripts & shell installers for classic MMOs. No subscription. No internet. Forever. Visibility: ✅ Public

Tracking Toolkit allows you to view and record OpenXR devices inside Blender with NLA integration and SMPTE timecode. It allows camera/object tracking with high degrees of clarity, and can also tra…

Apache-2.0 reference backbone wrapping NVIDIA's Kimodo motion model behind MMCP. Run Kimodo on your own GPU.

neural networks in pure C

Browser Harness | Self-healing harness that enables LLMs to complete any task.

runs anywhere. uses anything

Easy to use device for connecting "old" measuring units (water, power, gas, ...) to the digital world

Automatically enable wireless ADB and switch to port 5555 on Android device boot - no root required. Seamless wireless debugging on Chromecast with Google TV and other Android devices.

Posit Open Source Website

A state-of-the-art DirectX12 based pathtracer

HY-World 1.5: A Systematic Framework for Interactive World Modeling with Real-Time Latency and Geometric Consistency

Constraint-based geometry sketcher for blender

"O(log N) MoE Expert routing via RT Core ray tracing. BVH traversal replaces matrix multiplication in neural language models."

WebGPU physics engine based on the AVBD solver

TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Models

Dynamic context pruning plugin for OpenCode - intelligently manages conversation context to optimize token usage

Official implementation of Kimodo, a kinematic motion diffusion model for high-quality human(oid) motion generation.

The best Claude Code that $200 can buy

From-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) for LLM KV cache compression. 5x compression at 3-bit with 99.5% attention fidelity.