PulseAugur
EN
LIVE 08:25:12

LLMs run in-browser via WebGPU, cutting inference costs

By PulseAugur Editorial · [1 sources] ·

A new approach allows large language models to run directly within a web browser using WebGPU and Transformers.js, eliminating the need for server-side inference. This method, demonstrated in the BEINGS chatroom application, enables each user to run AI creatures on their own device, significantly reducing operational costs. Benchmarks show WebGPU achieving 63 tokens/s for a 360M-parameter model, with WASM as a fallback, though smaller models may exhibit reduced performance compared to hosted alternatives. AI

IMPACT Enables cost-effective deployment of LLMs in user-facing applications by shifting inference to the client.

RANK_REASON Demonstrates a novel technical approach for running LLMs client-side, but is not a release from a frontier lab.

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

LLMs run in-browser via WebGPU, cutting inference costs

How we ranked this

Signal score
35 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Demonstrates a novel technical approach for running LLMs client-side, but is not a release from a frontier lab.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
infra, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Matthew Charles Vladislav Busel ·

    Run an LLM in the browser: WebGPU did 63 tokens/s, WASM did 1.6

    <p><em>Originally published at <a href="https://tensorust.com/articles/run-llm-in-browser-webgpu/" rel="noopener noreferrer">tensorust.com</a>.</em></p> <p><strong>TL;DR:</strong> You can run a small language model entirely inside a web page with <a href="https://huggingface.co/d…

RELATED ENTITIES

RELATED TOPICS