Theta EdgeCloud has expanded its APIs and Model Context Protocol (MCP) server, allowing AI agents to independently find, deploy, and manage GPU infrastructure. An agent can now check available GPUs and prices, deploy a node, monitor its status, and review logs without a person using the platform dashboard. Theta EdgeCloud is a decentralized GPU infrastructure platform that connects computing resources with AI workloads.
The expanded API provides access to GPU availability, hardware specifications, regions and pricing, as well as tools for creating, starting, stopping, restarting, and deleting GPU nodes. The update also adds access to deployment events, logs, and read-only billing data. API keys are limited to a single project, so an automated system can only access the resources connected to that project.
MCP works as an additional layer above the API. It presents the available functions as tools that AI agents can call directly. This allows an agent, for example, to investigate a failed deployment by checking its status and logs or answer a billing question without being able to change account records.
*Image: *Magnific