GitHub - abi/secret-llama: Fully private LLM chatbot that runs entirely with a b...
source link: https://github.com/abi/secret-llama
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
Repository files navigation
Secret Llama
Entirely-in-browser, fully private LLM chatbot supporting Llama 3, Mistral and other open source models.
- Fully private = No conversation data ever leaves your computer
- Runs in the browser = No server needed and no install needed!
- Works offline
- Easy-to-use interface on par with ChatGPT, but for open source LLMs
Big thanks to the inference engine provided by webllm.
System Requirements
To run this, you need a modern browser with support for WebGPU. According to caniuse, WebGPU is supported on:
- Google Chrome
- Microsoft Edge
It's also available in Firefox, but it needs to be enabled manually through the dom.webgpu.enabled flag. Safari on MacOS also has experimental support for WebGPU which can be enabled through the WebGPU experimental feature.
In addition to WebGPU support, various models might have specific RAM requirements.
Try it out
You can try it here.
To compile the React code yourself, download the repo and then, run
yarn
yarn build-and-preview
If you're looking to make changes, run the development environment with live reload:
yarn
yarn dev
Supported models
Model | Model Size |
---|---|
TinyLlama-1.1B-Chat-v0.4-q4f32_1-1k | 600MB |
Llama-3-8B-Instruct-q4f16_1 ⭐ | 4.3GB |
Phi1.5-q4f16_1-1k | 1.2GB |
Mistral-7B-Instruct-v0.2-q4f16_1 ⭐ | 4GB |
Looking for contributors
We would love contributions to improve the interface, support more models, speed up initial model loading time and fix bugs.
Other Projects
Check out screenshot to code and Pico - AI-powered app builder
Recommend
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK