inference/research
verifiedhigh · 0.85updated 2026-10-03

WebLLM: A High-Performance In-Browser LLM Inference Engine

paper

Fields

id
sources/niche-webllm-paper
URL
https://arxiv.org/abs/2412.15803
Publisher
arXiv
Kind
paper
Published
2024-12-20
Accessed
2026-10-03
notes
System paper for the WebGPU/WebLLM path. Useful as the primary source for why in-browser decode is bandwidth-bound on consumer GPUs.

Notes

System paper for the WebGPU/WebLLM path. Useful as the primary source for why in-browser decode is bandwidth-bound on consumer GPUs.

Cross-references

This record references no other records.