Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,4 @@
|
|
|
|
|
| 1 |
title: Ministral WebGPU
|
| 2 |
emoji: ⚡️
|
| 3 |
colorFrom: red
|
|
@@ -9,43 +10,56 @@ short_description: Frontier multimodal AI, running entirely in your browser.
|
|
| 9 |
app_build_command: npm run build
|
| 10 |
app_file: dist/index.html
|
| 11 |
models:
|
| 12 |
-
|
| 13 |
-
|
|
|
|
|
|
|
| 14 |
|
|
|
|
| 15 |
|
| 16 |
-
|
| 17 |
-
|
|
|
|
|
|
|
| 18 |
|
| 19 |
-
Overview
|
| 20 |
This project demonstrates cutting-edge browser-based AI by running a complete 3B parameter multimodal language model entirely client-side using WebGPU acceleration. No servers, no API calls, no data sent anywhere - complete privacy and instant inference.
|
| 21 |
|
| 22 |
-
Key Features
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
This project showcases the future of AI deployment - moving powerful language models from cloud servers to the edge, where they can provide instant, private, and accessible intelligence without compromising user privacy or requiring expensive infrastructure.
|
| 41 |
|
| 42 |
-
Author
|
| 43 |
-
|
| 44 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
|
| 46 |
-
🎓 MSc Data Science & AI '25, Google WTM Scholar
|
| 47 |
-
🔬 Research areas: Computer Vision, NLP, Climate AI
|
| 48 |
-
💼 Experience: Gesture Recognition, Backend Development, ML Engineering
|
| 49 |
-
🔗 LinkedIn | GitHub | HuggingFace
|
| 50 |
-
License
|
| 51 |
Apache-2.0
|
|
|
|
| 1 |
+
---
|
| 2 |
title: Ministral WebGPU
|
| 3 |
emoji: ⚡️
|
| 4 |
colorFrom: red
|
|
|
|
| 10 |
app_build_command: npm run build
|
| 11 |
app_file: dist/index.html
|
| 12 |
models:
|
| 13 |
+
- mistralai/Ministral-3-3B-Instruct-2512-ONNX
|
| 14 |
+
- mistralai/Ministral-3-3B-Instruct-2512
|
| 15 |
+
---
|
| 16 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
| 17 |
|
| 18 |
+
# AI Multimodal WebGPU Assistant
|
| 19 |
|
| 20 |
+
**Developer:** Muhammad Abdullah Rasheed
|
| 21 |
+
**Research Assistant @ Cambridge | MSc Data Science & AI '25 | Google WTM Scholar**
|
| 22 |
+
|
| 23 |
+
## Overview
|
| 24 |
|
|
|
|
| 25 |
This project demonstrates cutting-edge browser-based AI by running a complete 3B parameter multimodal language model entirely client-side using WebGPU acceleration. No servers, no API calls, no data sent anywhere - complete privacy and instant inference.
|
| 26 |
|
| 27 |
+
## Key Features
|
| 28 |
+
|
| 29 |
+
- **Privacy-First Architecture**: The entire Ministral-3B model runs locally in your browser using WebGPU - your video feed never leaves your device
|
| 30 |
+
- **Real-Time Multimodal AI**: Live camera feed processing with visual question answering capabilities
|
| 31 |
+
- **WebGPU Acceleration**: Leveraging the latest browser GPU APIs for near-native performance
|
| 32 |
+
- **Zero Backend Dependencies**: No API keys, no server calls, no external services required
|
| 33 |
+
- **Cross-Platform**: Works seamlessly across modern browsers with WebGPU support
|
| 34 |
+
|
| 35 |
+
## Technical Stack
|
| 36 |
+
|
| 37 |
+
- **Model**: Ministral-3-3B-Instruct (quantized for browser deployment)
|
| 38 |
+
- **Runtime**: Transformers.js for in-browser inference
|
| 39 |
+
- **Compute**: WebGPU API for GPU acceleration
|
| 40 |
+
- **Frontend**: Modern JavaScript with WebAssembly integration
|
| 41 |
+
|
| 42 |
+
## Use Cases
|
| 43 |
+
|
| 44 |
+
- Visual question answering from live camera feed
|
| 45 |
+
- Real-time scene understanding and description
|
| 46 |
+
- Privacy-sensitive AI applications
|
| 47 |
+
- Edge computing demonstrations
|
| 48 |
+
- Educational tool for AI and browser technologies
|
| 49 |
+
|
| 50 |
+
## Why This Matters
|
| 51 |
+
|
| 52 |
This project showcases the future of AI deployment - moving powerful language models from cloud servers to the edge, where they can provide instant, private, and accessible intelligence without compromising user privacy or requiring expensive infrastructure.
|
| 53 |
|
| 54 |
+
## Author
|
| 55 |
+
|
| 56 |
+
**Muhammad Abdullah Rasheed**
|
| 57 |
+
Research Assistant | AI & Machine Learning Researcher
|
| 58 |
+
- 🎓 MSc Data Science & AI '25, Google WTM Scholar
|
| 59 |
+
- 🔬 Research areas: Computer Vision, NLP, Climate AI
|
| 60 |
+
- 💼 Experience: Gesture Recognition, Backend Development, ML Engineering
|
| 61 |
+
- 🔗 [LinkedIn](https://www.linkedin.com/in/muhammad-abdullahrasheed-/) | [GitHub](https://github.com/Abdullahrasheed45) | [HuggingFace](https://huggingface.co/Abdullahrasheed45)
|
| 62 |
+
|
| 63 |
+
## License
|
| 64 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
Apache-2.0
|