Coming VERY Soon: Distributed GPU Compute Across Devices
The SpawnDev.WebTorrent P2P network we're building creates a natural foundation for distributed GPU compute. Every connected device exchanges data over WebRTC — extending this to share compute workloads is the next step:
SpawnDev.ILGPU
AcceleratorType.P2P— 7th Backend — Distributes kernels across connected devices transparently. Same C# kernel code, sameLoadAutoGroupedStreamKernelAPI . The developer writes one kernel, it runs on 1 GPU or 10 GPUs across a household.SpawnDev.ILGPU.ML Model inference sharding — Split a 14B model across multiple devices. Each runs inference on their portion via SpawnDev.ILGPU, passes intermediate tensors to the next peer. A model that doesn't fit on one device runs across your phone, laptop, tablet, and desktop.
Volunteer compute pools — Users opt in to donate idle GPU time. Like Folding@Home for ML inference in the browser.
Every device in your home contributing to one shared AI compute pool — phone, laptop, tablet, desktop, old gaming PC. The living room becomes a compute cluster.
Resources
- SpawnDev.ILGPU — Run ILGPU C# kernels on WebGPU, WebGL, Wasm, Cuda, OpenCL, and CPU
- SpawnDev.ILGPU.ML — GPU ML inference + training for .NET
- SpawnDev.WebTorrent — Pure C# BitTorrent/WebTorrent for P2P model delivery
SpawnDev.ILGPU is ready to use NOW, 100%. SpawnDev.ILGPU.ML first official release, 4.0.0, is coming soon, possibly today. Distributed computing is our next tackle. Take a look. This is Ai for everyone, everywhere, on everything. 🖖
Top comments (0)