Hugging Face open-sources 207 WebGPU kernels for in-browser AI inference
nicodotdev · x · 2026-09-01
Hugging Face ships @huggingface/kernels: 207 open-source WebGPU kernels loadable straight from the Hub. The author notes fast browser inference is a multi-layer effort — browser-friendly model formats, runtimes with efficient execution plans, and GPU kernels tuned for the browser — and this release provides the bottom layer.
Related event: Hugging Face open-sources 207 WebGPU kernels for faster browser inference(2 posts)→
More from Infra
- SGLang Summit sponsors grow from 11 to 36 in one week; Lip-Bu Tan, Lilian Weng to speak — BanghuaZ · 2026-10-18
- Yacine jokes his datacenter KPI is gigawatt-hours: get paged if not burning enough power — yacineMTB · 2026-10-18
- 1.68x inference speedup drops, directly accelerating GRPO post-training — yacineMTB · 2026-10-18
- OpenRouter is only a fraction of the total addressable market, poster argues — sloppenheimer · 2026-10-18
- Developer builds change alerts for LLM prices and retirements across Bedrock, Azure and Vertex — jsandhu0780 · 2026-10-18
- 12.3M tokens in 27 minutes for $0.14: agent session with 99.6% cache hit — teortaxesTex · 2026-10-18