nz.gpu/gpu-cloud: Model Context Protocol (MCP) Server
The nz.gpu/gpu-cloud MCP server provides access to renting cloud GPUs for AI agents. It supports live pricing and VRAM estimates, along with operational controls to launch GPU instances, view logs, and stop running sessions. It is described as focused on GPU provisioning with real-time details.
🛠️ Key Features
- Live prices for cloud GPU rentals
- VRAM estimates
- Launch GPU instances
- View logs
- Stop GPU instances
🚀 Use Cases
- Run AI workloads on rented cloud GPUs
- Select a GPU tier based on estimated VRAM
- Monitor execution via logs and terminate via stop
⚡ Developer Benefits
- Programmatic GPU launch, logging access, and shutdown
- Up-to-date pricing and VRAM expectations for agent workflows
⚠️ Limitations
- Limited to the capabilities described: pricing, VRAM estimates, launch, logs, and stop.