Project Quiver documentation, designs, and project artifacts.

Trackers gives you clean, modular re-implementations of leading multi-object tracking algorithms released under the permissive Apache 2.0 license. You combine them with any detection model you alre…

A fully open-source humanoid arm for physical AI research and deployment in contact-rich environments.

Single-image to 200-view splat-training dataset (TRELLIS.2 + Pixal3D) with empirical 5-tier sampler system, probe mode, autocrop + AuraSR preprocessing, hot-folder daemon with Windows BAT integration.

Full Transformer into a custom chip. microGPT in RTL, generating names on a Virtex-5 FPGA at ~56k tokens/second.

HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videos

Official source code of FreeCAD, a free and opensource multiplatform 3D parametric modeler.

A library of agent skills for CAD, CAE and CAM

HRM-Text is a 1B text generation model based on the HRM architecture, strengthened by task completion and latent space reasoning.

microGPT benchmarks: a single M4 Max MacBook Pro P-core in C runs Karpathy's 4192-parameter transformer at ~71x the throughput of TALOS-V2's FPGA implementation.

OSS Legal AI Platform

Code, 3D Print Files, BOM and Build Instructions for the DIYNAFLUOR DNA Fluorometer

PicoGK is a compact and robust geometry kernel for Computational Engineering

Reverse of OpenAI Privacy Filter: same 1.5B model, returns PII as structured spans instead of masking.

1K resolution vision transformers pretrained on 1B human images.

A minimal PyTorch re-implementation of AlphaFold2's model & training

Instant, Concurrent, Secure & Lightweight Sandbox for AI Agents.

A complete private computer on a USB stick. PAI is a bootable Debian 12 Linux distro that runs Ollama locally on any x86_64 or ARM64 machine — full desktop, local AI, hardened networking, all in RA…

The Geometry of Forgetting

LLM speculative inference server for consumer hardware & heterogeneous computing

Tensor library for machine learning

Benchmarking mask interpolation methods for real-time on-device video object segmentation on Apple Silicon. 11 methods compared (optical flow, optimal transport, baselines) across 30K+ evaluations …

Fast state-of-the-art image and video segmentation in portable C/C++

A retargetable MLIR-based machine learning compiler and runtime toolkit.

A Spitting Image: Modular Superpixel Tokenization in Vision Transformers

SpectralQuant: Calibrated Eigenbasis Rotation and Water-Filled Bit Allocation for KV-Cache Compression