The deterministic tax engine for AI agents. Highest score ever recorded on TaxCalcBench (96% exact returns)

Fast Tokens

Language model tokenization at GB/s

A hive mind communication platform

Think with AI beyond the chat box. A shared canvas for handwriting, equations, diagrams, and spatial reasoning.

All MYVU clients in one monorepo

Starter Template for your Digital Doppelgänger.

Multi-slot LLM inference on AMD Strix Halo: recipes + honest benchmarks (236 tok/s @ 32 streams, llama.cpp Vulkan)

High-performance single-GPU inference for selected model checkpoints and GPUs.

Machine-readable map of the computational frontier around Erdős prize problems: 95 audited (17 board-ready), walls named, verifier-first P42 board specs. No Erdős prize is claimable by finite compu…

🐎 Ryan Lopopolo’s anthology, field guide, and agent context bundle for harness engineering

Pure Rust Inference Engine

A full GUI experience on top of llama.cpp: all-knobs model tuning, one-click build/update from upstream, HuggingFace discovery with VRAM-fit ratings

Adaptive tiered KV cache management for LLM inference. Single-controller, 8-number policy surface, fault-tolerant migration engine. Hardware-agnostic.

See-through-walls multiplayer AR built on MultiSet VPS. Use smartphones and rayban glasses to "see" tracked entities through walls.

God's Eye View v1 public open-source release placeholder

Reference code for the Meta-Harness paper.

LLM inference in C/C++

Local AI setup for AMD Strix Halo APU - Lemonade + Vulkan + kyuz0

Qwen3.6-35B-A3B-heretic NVFP4 + DFlash speculative decoding on DGX Spark (GB10/sm_121a). Source-built vLLM image + 7 patches + comprehensive deployment guide.

Your Computer. Anywhere.

ROCmFPX Family for AMD Hardware and Processors. More quants and special agent quants

Auto-resume Claude Code, Codex CLI, Grok, Qwen Code, Kimi CLI, OpenCode & Antigravity when the 5-hour/weekly usage limit resets — tmux, Zellij, VS Code & desktop apps

Quasar is an inference engine for Qwen3.6-27B-MTP on a single RTX 5090, in the spirit of antirez/ds4

Companion code for the global workspace interpretability paper

Interactive Jacobian-Lens visualizer and live steerer for GGUF models on llama.cpp

Benchmark to measure what the real knowledge cutoff of a model is

High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.

SteamOS-like ARM handheld distro