Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI
Source
huggingface.co
Date
Terms in this piece · Glossary
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
Why it matters
Browser-side inferenceRunning a trained model to get answers — the phase where AI is actually used, as opposed to trained.Full definition → gets a shared, versioned kernel layer instead of per-project shader code, and Fleet gathers correctness and performance evidence from hardware you could never test yourself.
