nano-chat-ui

KO | EN

gitlite — search

#hacktoberfest #ai #typescript #llm #go #deep-learning #nodejs #macos #java #cli #vue #pytorch #mcp #openai

(★ 9)

A lightweight, client-side, zero-install HTML-based chat interface. It enables you to communicate with various local and cloud-based AI providers (via OpenAI compatible APIs like Ollama, LM Studio, etc.) alongside Chrome's built-in local model (Gemini Nano).

nano-chat-ui Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Nano Chat UI

A lightweight, client-side, zero-install HTML-based chat interface. It enables you to communicate with various local and cloud-based AI providers (via OpenAI compatible APIs like Ollama, LM Studio, etc.) alongside Chrome's built-in local model (Gemini Nano).

![Nano Chat UI - Screenshot 1](https://raw.githubusercontent.com/ixabolfazl/nano-chat-ui/main/image.png)
![Nano Chat UI - Screenshot 2](https://raw.githubusercontent.com/ixabolfazl/nano-chat-ui/main/image2.png)

## Features

- **Multi-Provider Support**: Easily add and configure any OpenAI Compatible API endpoint (e.g., LM Studio, Ollama, OpenAI).
- **Chrome Built-in AI**: Support for local Chrome Gemini Nano via the experimental Prompt API.
- **Dynamic Configuration**: Customize system prompts, Temperature, and Top-K values for each conversation.
- **History Management**: Automatically stores chat sessions and providers locally using `localStorage`. Support for importing/exporting conversations and providers list.
- **RTL & LTR Support**: Automatically toggles language layout direction between Persian (FA) and English (EN).
- **Single File**: Zero installation or local dependencies required. Just open the HTML file.

## How to Use

1. Download the `nano-chat-ui.html` file from this repository.
2. Double-click to open `nano-chat-ui.html` in any modern web browser.
3. **To Connect to Local Providers (like Ollama or LM Studio):**
   - Click the gear icon (**Settings**) in the top bar.
   - Go to the **Providers** tab and click **Add Provider**.
   - Fill in a name and the **Base URL** (e.g., `http://localhost:11434/v1` for Ollama, or `http://localhost:1234/v1` for LM Studio).
   - Click **Test Connection** to automatically fetch available models.
   - Check/uncheck models to configure their visibility and click **Save**.
4. **To enable local Chrome Gemini Nano:**
   - Ensure you are on a compatible Google Chrome version with Chrome AI flags enabled (`chrome://flags/#optimization-guide-on-device-model` set to Enabled, and `chrome://flags/#prompt-api-for-gemini-nano` set to Enabled).

> **Note on CORS**: Connecting to local ports from browser-hosted pages requires CORS to be enabled on your model provider. For Ollama, start it with the environment variable set: `OLLAMA_ORIGINS=*`.

nano-chat-ui

nano-chat-ui Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

nano-chat-ui

nano-chat-ui Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README