AI Assistant

Initializing AI Assistant

Checking WebGPU support...

First load downloads ~200MB model (cached after)

[VISION_FEEDS]

[WEBCAM]

> CLICK TO ACTIVATE

OFFLINE

[SCREEN]

> CLICK TO SHARE

OFFLINE

📁 Upload Image

⏱ Capture Rate

1 frame/sec

0.5 5

0 frames captured

[VISION_CONTEXT]

No frame captured

Descriptions will appear here...

0 observations

Running locally in your browser with WebGPU

Try asking:

Hold Space to talk (click input first to type)

PROCESSING...

Ctrl+Enter Send Esc Cancel Ctrl+N New Chat

[SYSTEM]

AI Model

Provider

Ready

Vision Model

Ready

Model Settings

Context Window: 8K

Max Output: 512 tokens

Session

Messages: 0

Visual Frames: 0

Performance

Speed: -- tok/s

Backend: Detecting...

Total Tokens: 0

Voice

Auto-speak responses

TTS Engine

Ready

Memory

Web Search

Get free API key →

No key configured

Thinking Mode

Deep Thinking

Step-by-step reasoning

Research Mode

Auto-Research

Auto-detect complex queries and perform multi-step research

About

This AI runs entirely in your browser using WebGPU. No data is sent to servers.

v1.0.0

Initializing AI Assistant

Initialization Failed

AI Assistant