GaiaNet

Introducing the Gaianet AI network

Getting started running a GaiaNet node on your personal computer!

GaiaNet node

Each GaiaNet node provides a specialized API service that encapsulates a unique combination of

• a specialized and fine-tuned LLM (e.g., an LLM that excels in answering questions about the Rust programming language)
• a domain-specific knowledge base (e.g., knowledge about the WasmEdge project)
• an inference app that manages the context and history of conversations (e.g., RAG and MemGPT prompt injection)
• compute resources required to run the LLM app as an API service (e.g., a Nvidia GPU or a Mac M3 device)

The GaiaNet node API service is fully compatible with the OpenAI JSON spec, and hence each GaiaNet node can function as an alternative backend for OpenAI-based frontends or agents.

GaiaNet domain

GaiaNet nodes are organized into GaiaNet domains so that they can be discovered and accessed from the world. A GaiaNet domain groups together nodes that have similiar purposes, and provides a single Internet domain for all its node agents. For example, agents that answer student questions at University of California at Berkeley could be organized under the gaianet.berkeley.edu domain. The doman owner vouches for its node agents by puttings its reputation behind the agents. Hence, a staking mechnism is designed for the domains. If a domain is found to have too many inactive or even misbehaving agents, the domain stakers could be slashed.

GaiaNet protocol

The GaiaNet protocol connects and incentivizes GaiaNet nodes and domains to form a coherent network of web services for AI agents. It provides a mechanism to discover, connect to, and pay for GaiaNet node services through a decentralized marketplace. It also incentivizes domains to manage node agents through a staking program. Furthermore, the GaiaNet protocol connects model creators (i.e., people who have skills to finetune models) and knowledge providers to node operators through a marketplace.

Join us!

• Website
• Twitter
• Discord
• Telegram
• Medium