Play Hub
Last updated
Last updated
The Play Hub serves as the deployment and reward mechanism for AI agents. Models created across the Training Grounds are deployed on the AI agents and model marketplace, utilizing centralized or decentralized compute to evaluate outputs. Developers and applications can integrate these agents using the Play AI SDK, which overlays existing systems to enhance functionality. Developers are incentivized based on the usage of agents and models, while also get ranked and rewarded by the network.
The Play AI SDK deploys the agents and models across the marketplace.
Developers can use the SDK to integrate AI agents into various existing applications.
An intent for a specific AI agent is generated by the client application and processed by the intent operator.
Nodes select and deploy the appropriate agent from the marketplace for inference.
Inference occurs via centralized GPU providers, decentralized platforms, or nodes.
The results are streamed into the application and to a nodes/benchmarking unit for evaluation.
Scores and rankings are generated based on predefined criteria, and nodes are rewarded or penalized based on their contributions to the agent's performance.