An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs

Skills for agents to view markets and place trades on Gensyn's Delphi prediction markets

Gensyn Reproducible Execution Environment (REE) enables fully reproducible execution of machine-learning workloads.

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

Build, run, and manage agent platforms.

It's a TUI game... that's it

BlockAssist is an AI assistant that learns from its user’s actions in Minecraft.

open-source coding agent

[NeurIPS 2025 Spotlight] Reasoning Environments for Reinforcement Learning with Verifiable Rewards

Official code for the All is not Lost paper

Collective communications library with various primitives for multi-machine training.

A tool for gathering and retaining your own Whoop data.

Copybara: A tool for transforming and moving code between repositories.

A fully open source framework for creating RL training swarms over the internet.

Ensures bitwise reproducibility of ML operations across different hardware targets.

Accommodating Large Language Model Training over Heterogeneous Environment.

Get up and running with Kimi-K2.6, GLM-5.1, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Distributed Training Over-The-Internet

Distributed Peer-to-Peer Web Search Engine and Intranet Search Appliance

Debloated fork of the extension "I don't care about cookies"

YaFSDP: Yet another Fully Sharded Data Parallel

Streamlining reinforcement learning with RLOps. State-of-the-art RL algorithms and tools, with 10x faster training through evolutionary hyperparameter optimization.

A high-throughput and memory-efficient inference and serving engine for LLMs

PDF Reader in JavaScript

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

Protect your data against global mass surveillance programs.

Convert ONNX models to PyTorch.

The simplest and most extensible zkVM. Fast and fully open source from a16z crypto and friends. ⚡