The fastest way to run Qwen 3.8 Flash Next, Qwen 3.8 27B and Ternary Bonsai 2 27B on a Mac: 125 tok/s in OpenCode on an M5 Max, and a 27B model on 16 GB Macs. Native MTP speculative decoding on App…

Open source inference engine for agents that optimizes itself for your exact hardware. Compiles and tunes its kernels on your device, so open models run up to 2x faster than llama.cpp. Works on App…

Millisecond decisions, any domain: a 0.8B open "System 1" model that picks between your options with calibrated probabilities. One base, swappable LoRA adapters, on your own hardware.

PDF Reader in JavaScript

Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.

Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

Bend 2: a fast language that blocks AI mistakes via proof. Install: curl -fsSL https://bend-lang.com/install.sh | sh

Local-first desktop markdown editor

The fastest grep in the world. (BEING PORTED TO A [secret] LANGUAGE '...' AS OF NOW -mtyrkba 8 October))

A decompilation of Super Smash Bros Melee brought to you by a bunch of clever folks.

The beautiful & flexible React.js docs framework.

Create polished demo videos without editing skills. Mac/Windows/Linux

Build a system, load it until it breaks, and watch why. A system design simulator for learning distributed systems.

Tensor library for machine learning

Convert Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF to clean Markdown. Built in Rust, with Node.js and Python bindings.

Inspect, annotate, and give feedback on any live interface.

meat.dev

Open Frontier Intelligence

the runtime your coding agents live on

Fast WebGPU runtime for 1-bit (binary-weight) LLMs in the browser. Bit-exact, zero runtime dependencies.

Fast and accurate AI powered file content types detection

SpaceXAI's coding agent harness and TUI. Fullscreen, mouse interactive, extensible.

Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

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

An open-source protocol enabling high-performance, local-first, peer-to-peer, end-to-end encrypted collaborative apps.

An app to monitor the (Codex) situation

DuckDB is an analytical in-process SQL database management system

Self-hosted AI workspace.

A fast, helpful, and open-source document parser