Hugging Face ships 207 open-source WebGPU kernels to run AI models directly in the browser
Hugging Face · youtube · 2026-09-04
Hugging Face has open-sourced 207 WebGPU kernels for browser-based AI on the Hub, along with the @huggingface/kernels library that loads, validates, renders and runs them. The accompanying video explains the design: the difference between ops and kernels, why WGSL, and why kernels ship as Jinja templates instead of fixed WGSL files — the browser picks supported data types and workgroup sizes to build the fastest kernel for each device. Demos include an attention mechanism using tinyBERT weights in 20 lines of JavaScript and a 1M+ cell wave animation driven by a single matmul kernel. The new Fleet platform lets users run kernel benchmarks on their own GPUs to help optimize for untested devices.
More from coding & agent
- Dimillian builds a game from scratch in an hour with Astra Ultra, using just 5% of his weekly quota — Dimillian · 2026-09-05
- Full walkthrough: how Nissim Lebovits used AI agents to find 3 million missing people — MaxLenormand · 2026-09-05
- Cortex: an MIT-licensed local SQLite memory layer for coding agents, open-sourced for feedback — Royal_Philosopher_58 · 2026-09-05
- Dev builds unified directory indexing AI agents, MCP servers and skills — exposed as an MCP server itself — No_Cake8366 · 2026-09-05
- Bug Hunt Bench: 105 real bugs stress-test GPT-6, Claude, Grok, Gemini and more coding agents — PawelHuryn · 2026-09-05
- Hermes Agent adds Perplexity as selectable web search and scraping backend — Teknium · 2026-09-05