Start Here: The Protocol for Independent Ai

We are on a mission to make Independent Ai the global standard. How does 6079 establish standards for Decentralized Ai? You can think of 6079 as the connective tissue of a much larger decentralized Ai body. 6079 co-develops standards to help connect every part of the Decentralized Ai stack all the way from GPU providers and aggregators all the way to the agents that consume them. The advantages of decentralized protocols come from their open, free market, permissionless interoperability—these standards are not yet in place for this fast growing industry. The 6079 protocol secures Decentralized Ai Agents and GPU Networks with cryptoeconomic security and interoperability standards through the following three pronged approach:

  1. Industry Consortium: Organize and align a governing body of leading De/AI projects, protocols, and contributors toward open and permissionless interoperability

  2. Community Grants: Build the protocol by funding the development of the protocol and open standards

  3. Proof of Inference Standard: Secure Ai compute through cryptoeconomic protocol for validating gpu compute

These three efforts contribute to a virtuous evolutionary cycle that 1) gathers and organizes project, protocols, companies, and contributors who are building decentralized Ai 2) Provides a grant mechanism for contributions to the protocol 3) Secures the network by operating the 6079 Proof of Inference Protocol.

Industry Consortium

6079 is much more than a protocol—it is a leader in a decentralized movement to make Independent AI the global standard. The consortium organizes and aligns industry leading protocols, projects, and contributors in the movement.

There is a massive opportunity in DePIN networks to compete with centralized systems to become the standard in the Ai Agent economy which is expected to reach $4.4 trillion annually [*], this opportunity presents a growing incentive for bad actors to find ways to extract value from these networks. It is in the best interest of the industry and all users to cooperate to build shared standards to increase the competitiveness of DePIN networks against alternatives. A decentralized group of ideologically aligned projects, protocols, and companies are supporting this effort.

We propose an open source industry standard framework for ensuring that connected GPU hardware clusters:

  1. Maintain compliant hardware connections at the performance level claimed

  2. Maintain the right data through inference with maximum privacy options

  3. Hold the correct model weights in gpu memory

  4. Achieve cryptoeconomic security

Goals:

  1. Create a public foundation of trust in Decentralized Ai

  2. Share a standard for verification of Ai inference on decentralized GPUs

Guiding Principles:

  • Open Source

  • Decentralized

  • Composable

  • Interoperable

  • Protocol Governed

Participating Projects:

Announcing Next Week

Gamified Proof of Inference

The Proof of Inference Protocol (PoIP), is a system for enabling effective peer-to-peer AI compute within an open and decentralized AI ecosystem. The protocol uses competitive game mechanics to put Ai endpoints into direct competition and demonstrate their capabilities such as latency, token throughput, processing power, reputation, etc. There are primary actors in this system:

  1. Providers: Provide an API endpoint for a specific model, running on specific hardware

  2. Validators: Coordinate as a swarm to run POIP protocol software, and validate providers

  3. Agents: Act as the third party actor making requests to validators

How does it work?

Validators stay in constant communication through a distributed hash table. They follow the protocol instructions to select Provider end points and run game simulations that put the end points in direct competition.

Games are designed to be undetectable to the Providers and test various aspects of Ai inference including: latency, throughput, model type, hardware capabilities, etc. New game simulations can be created and implemented by the consortium.

Validators, Providers, and agents are granted $THINK tokens when successfully completing their epoch. A staking/slashing mechanism will be introduced to further increase cryptoeconomic security.

Contributor Grants

Individuals and Companies can participate in working groups that contribute to the movement and protocol. The consortium is creating working groups that define objectives, strategy, and projects to evolve the protocol and standards. These working groups are assigned token budgets weighted to the work and impact involved in a project and are awarded with $THINK governance tokens.

Initial working groups:

  • Identity, Transactions, and Private Messaging v2

  • Scheduler, Attribution, and Reputation v2

  • Agent interface standardization v2

  • Open source inference engine V2

  • Season 2 Community Campaign

  • Contributor organization

Last updated