WebGPT

Visit Tool

WebGPT is a Coding & Development tool that runs GPT models directly in the browser using WebGPU. It serves as a proof-of-concept and educational resource for developers, implemented in less than 1500 lines of vanilla JavaScript.

Claim this tool

2Views

At a glance

Pricing

Open Source

Free tier

Yes

API

Skill level

Technical

About

What is WebGPT?

WebGPT is an innovative project demonstrating the capability to run GPT models directly within a web browser, leveraging the power of WebGPU. This implementation, crafted in under 1500 lines of vanilla JavaScript and HTML, functions as both a proof-of-concept and an educational resource for developers interested in on-device AI inference. It has been successfully tested with models up to 500 million parameters, with potential for larger models through further optimization. The project highlights the significant advancement WebGPU brings to web applications, offering near-native access to the GPU and compute shaders. Developers can easily run WebGPT by cloning the repository and using a compatible browser like Chrome Canary or Edge Canary, with options to use included models or import custom ones.

Best used for

Ideal for developers and students who need to understand the practical implementation of GPT models, experiment with WebGPU for on-device AI inference, and learn about transformer architecture. Especially valuable for those looking for a lightweight, vanilla JavaScript example of AI in the browser.

Common actions

run AI models

learn WebGPU

implement transformers

experiment with GPT

deepfakelow-code/no-codeautomated workflowworkflowscollaborationface swapping"AI Agents"open-sourcegithub copilot

Capabilities

Key features

Run GPT in browser
Utilizes WebGPU
Vanilla JavaScript
Supports custom models
Educational resource

Target Audience

studentteacher

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What are the system requirements to run WebGPT?

To run WebGPT, you need a compatible web browser that supports WebGPU, such as Chrome v113 or newer, or Chrome Canary/Edge Canary. It's recommended to clone the repository and run it locally, as loading weights remotely can be significantly slower.

Can I use my own custom GPT models with WebGPT?

Yes, WebGPT supports importing custom models. The project includes conversion scripts in the 'misc/conversion_scripts' directory to help you prepare your models for use with the WebGPT framework. This allows for flexible experimentation.

What is the performance of WebGPT on different hardware?

Performance varies by hardware and model size. For example, on a 2020 M1 Mac, it achieves 3ms/token at 5M parameters and 120ms/token at 775M parameters with f32 precision. Larger models like 1.5B parameters are functional but currently less stable.

Trending

Subcategories trending in Coding & Development

Open Source & Models Code Assistants DevOps & Infrastructure No-Code / Low-Code Testing & QA Backend & APIs

Trending

Also listed in

This tool also appears in

Research & Education › Study Assistants

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce