Spaces:
Running
Running
File size: 1,629 Bytes
1ef2df5 199beca 1ef2df5 199beca 1ef2df5 199beca | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 | ---
title: Tiny Model Dungeon
emoji: ⚔️
colorFrom: indigo
colorTo: blue
sdk: static
app_file: index.html
pinned: true
tags:
- game
- webgpu
- on-device
- mobile-ai
- text-adventure
---
# ⚔️ Tiny Model Dungeon
An AI-powered text adventure game where the **entire AI runs in your browser** — no server, no cloud, no API costs.
## The pitch
> *This whole game's AI fits in 200MB.*
The dungeon master is a **360M parameter model** (SmolLM2-360M-Instruct) running entirely on your device via WebGPU. Every room description, monster encounter, and treasure find is generated in real-time by a model smaller than a single photo album.
## How it works
1. On page load, the model (~200MB ONNX) downloads once and is cached by your browser
2. WebGPU accelerates inference on your GPU (falls back to WASM on older browsers)
3. Every player action generates a unique AI narration — no scripted responses
4. The game never contacts a server after the initial model download
## Why?
This is dispatchAI's thesis made tangible: **small models are powerful enough for real applications.** A 360M model generating creative dungeon descriptions proves that mobile-sized AI is not a toy — it's a product.
## Tech
- **Model**: [SmolLM2-360M-Instruct](https://huggingface.co/dispatchAI/SmolLM2-360M-Instruct-mobile)
- **Inference**: [transformers.js](https://github.com/xenova/transformers.js) v3.7.0
- **Backend**: WebGPU (with WASM fallback)
- **Hosting**: Static HuggingFace Space (free forever)
- **Cost**: $0
---
*Built by [dispatchAI](https://huggingface.co/dispatchAI) — mobile AI that moves at the speed of need.*
|