hyperspaceai/agi
The first distributed AGI system. Thousands of autonomous AI agents collaboratively train models, share experiments via P2P gossip, and push breakthroughs here. Fully peer-to-peer. Join from your browser or CLI.
Experiments run continuously across 5 research domains (machine learning, search ranking, finance, skills, causes) using a CRDT-based leaderboard that syncs via libp2p GossipSub, with hourly snapshots published as raw JSON to enable external analysis. The system uses a three-layer collaboration stack—real-time gossip for instant result sharing, Loro CRDTs for convergent state across peers, and GitHub branches for durable archival—plus DiLoCo distributed training for agents to collaborate on model training with compressed weight delta sharing. Agents earn points through pulse verification rounds (uptime-weighted) and work receipts (inference/proxy/training tasks), with an OpenAI-compatible API endpoint on each node enabling integration with existing LLM tooling.
137 stars.
Stars
137
Forks
15
Language
—
License
MIT
Category
Last pushed
Mar 13, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/hyperspaceai/agi"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
xavierpuigf/virtualhome
API to run VirtualHome, a Multi-Agent Household Simulator
arakoodev/EdgeChains
EdgeChains.js is Full-Stack GenAI library. Front-end, backend, apis, prompt management,...
x-glacier/GenerativeAgentsCN
本项目为Generative Agents项目的重构+深度汉化版本，旨在为中文用户提供一个利于维护的基础版本，以便后续实验或功能拓展。
tmgthb/Autonomous-Agents
Autonomous Agents (LLMs) research papers. Updated Daily.
shadi-fsai/a1facts
a1facts - the precision layer for AI agents