🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading

Decentralized deep learning in PyTorch. Built to train models on thousands of volunteers across the world.

A crowdsourced distributed cluster for AI art and text generation

💬 Chatbot web app + HTTP and Websocket endpoints for LLM inference with the Petals client

The default client software for volunteering to generate images for the crowdsourced AI-Horde

📜 A python library for distributed training of a Transformer neural network across the Internet to solve the Running Key Cipher, widely known in the field of Cryptography.

Distributed AI research platform. Volunteer compute for autonomous ML experimentation. SETI@home meets autoresearch.

A Docker-based volunteer computing platform, developed during Mei-Chu Hackathon 2015 at NCTU.

Open-source SETI candidate review network for BYOK API workers and Codex/Claude Code agent clients.

♻️ Donate your idle AI compute. A marketplace of vibe-coding projects + a pool of donated tokens — OpenAI-compatible proxy in front, volunteer nodes (Claude/GPT/Grok/Ollama) behind. MIT.

Distributed peer-to-peer LLM inference network. Volunteer your GPU, earn AI credits, run any open-source model for free. Anonymous, encrypted, unstoppable.

Semantic coherence verification for distributed AI inference. Omega metric validated AUC 0.9539 on Wikipedia ES.

Democratize AI: a download-and-forget app that turns idle computers into a language model owned by everyone. Folding@home, pointed at intelligence instead of proteins.

AuspexAI volunteer worker for donating compute to AI research

An open, volunteer-powered, rule-governed AI network with a guaranteed kill-switch — a tool-using assistant, a federated model that survives any node going offline, and bounded self-improvement. AGPL-3.0.

Performing deep learning analysis on Video Stream using Volunteer Computing. Volunteer computing is a form of distributed computing where every client can join and leave at will and request processing. Each client will contribute a certain amount of its resources.

🚀 Accelerate LLM training with Fast-LLM, an open-source library for high-speed, scalable, and flexible model development.

AGILLM4.1 mainline single-file Transformer with DiffusionBlocks, AR/SAT/NAT heads, async side workers, and staged inference