🏠 Fully Client-Side Chat Over Documents 🏠

🏡Yes, it's another chat over documents implementation... but this one is entirely local!
🌐The vector store (Voy) and embeddings (Transformers.js) are served via Vercel Edge function and run fully in the browser with no setup required.
⚙️
The default LLM is Mistral run locally by Ollama. You'll need to install the Ollama desktop app and run the following commands to give this site access to the locally running model:
```
$ OLLAMA_ORIGINS=https://webml-demo.vercel.app OLLAMA_HOST=127.0.0.1:11435 ollama serve
```
Then, in another window:
```
$ OLLAMA_HOST=127.0.0.1:11435 ollama pull mistral
```
🗺️
The default embeddings are Nomic Embed v1. For more speed on some machines, switch to
```
"Xenova/all-MiniLM-L6-v2"
```
in
```
app/worker.ts
```
.
🦜LangChain.js handles orchestration and ties everything together!
🐙This template is open source - you can see the source code and deploy your own version from the GitHub repo!
👇Try embedding a PDF below, then asking questions! You can even turn off your WiFi.