Browser‑Based AI for Production: Practical Use Cases

Описание и применение

Маленькие модели, работающие в браузере, можно использовать в продакшене: удаление фона, генерация альтернативного текста, создание тикетов и семантический поиск без внешних API.

Задача

Can small AI models run directly in the browser be useful for real production applications?

Результат

The author proved that small, locally‑running AI models can deliver useful features in production, such as background removal, alt‑text generation, structured ticket creation, and semantic search, all without external API calls.

Рабочий процесс

• 1. Set up a React/TypeScript app with Web Workers for inference.
• 2. Load models on demand via Transformers.js (WebGPU or WebAssembly fallback).
• 3. Implement background removal with RMBG‑1.4 (≈170 MB).
• 4. Implement alt‑text generation with Florence‑2‑base‑ft (≈350–450 MB).
• 5. Implement smart ticket creation with Qwen2.5‑0.5B‑Instruct (≈786 MB).
• 6. Implement semantic search with all‑MiniLM‑L6‑v2 (≈23–90 MB).
• 7. Validate structured outputs with Zod and provide editable UI.
• 8. Deploy live demo and host code on GitHub.

Доказательства

I wanted to find out how we could use browser-based AI today, right now, in real production applications.

Источник

The models are loaded on demand, so you don't need to download all five of them just to open the application.

Источник

RMBG-1.4, with approximately 44 million parameters, does a really impressive job of separating an object from its background.

Источник

Florence-2-base-ft, a vision model with approximately 230 million parameters, to generate descriptions of images.

Источник

Qwen2.5-0.5B-Instruct, a small language model with approximately 500 million parameters.

Источник

all-MiniLM-L6-v2, a small embedding model with approximately 22 million parameters.

Источник