# mlc-ai/web-stable-diffusion

Bringing stable diffusion models to web browsers. Everything runs inside the browser with no server support.

Repository: https://github.com/mlc-ai/web-stable-diffusion
Canonical: https://ross.abutalabs.com/products/web-stable-diffusion
Homepage: https://mlc.ai/web-stable-diffusion
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Topics: webgpu, deep-learning, stable-diffusion, web-assembly, webml, tvm
Last push: 2024-03-12T13:29:56+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 91
- inputs: {"age_days": 1277, "days_push": 904, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3721, forks 238 (observed 2026-08-28T04:08:15.512402+00:00)

## What it is
A project that compiles and runs Stable Diffusion text-to-image models entirely inside web browsers using WebGPU and WebAssembly, with no server-side inference. It provides a Python-first MLC/TVM-based workflow for optimizing and deploying the models to the browser.

## Use cases
- generate images from text prompts in the browser
- run stable diffusion locally without a GPU server
- build privacy-preserving client-side AI image generation web apps
- reduce inference hosting costs by offloading compute to the client
- experiment with WebGPU machine learning in the browser
- compile and optimize diffusion models for web deployment

## When to choose
- you want fully client-side text-to-image generation with no backend
- privacy matters and prompts/images should never leave the user's device
- you want to cut server GPU costs for image generation
- you're building a WebGPU/WebAssembly ML demo or web app

## When to avoid
- you need broad browser support - WebGPU is not universally available
- you need the fastest or highest-quality generation on powerful server GPUs
- you need a production-ready, actively evolving stable diffusion toolkit
- your target users have low-end devices with limited memory

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, gpu-computing, compiler
- domain: artificial-intelligence, image-processing, web-development, deep-learning
- platform: browser, wasm, python
- tags: webgpu, stable-diffusion, text-to-image, webassembly, tvm, client-side-inference, mlc, web-server

## Member repositories
- mlc-ai/web-stable-diffusion (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:15.512402+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:29:19.579275+00:00, confidence not recorded.
  - readme: https://github.com/mlc-ai/web-stable-diffusion (fetched 2026-08-28T04:08:15.512402+00:00, sha 094b3658e02c)
  - homepage: https://mlc.ai/web-stable-diffusion (fetched 2026-08-29T09:24:01.853707+00:00, sha c4554b7f1088)
  - registry_pypi: https://pypi.org/pypi/web-stable-diffusion/json (fetched 2026-08-29T09:24:01.863807+00:00, sha cc265bfd826f)
- Data as of 2026-08-30T08:39:29.467469+00:00.