Initializing, please wait a moment

Everything runs on your device. Your messages are not uploaded and there is no account. The model downloads once (hundreds of MB) the first time and is cached by your browser for next time. A WebGPU browser is required.

Private AI Chat


Chat with an AI model that runs entirely on your device, in your browser. It suits quick drafting, brainstorming, or asking a question you would rather keep off a server.

The AI model runs in your browser on your device - messages are not uploaded
Pick a model and chat on your device; nothing is uploaded.

How it works

Pick a model and click Load model. The first time, the model downloads to your browser (hundreds of MB) and is cached for next time; after that it runs entirely on your device. Your messages are never uploaded, there is no account, and nothing is sent to a server. A WebGPU browser is required - the latest Chrome or Edge on desktop work well.


Choosing a model

Private AI Chat - four available models with download sizes: TinyLlama 1.1B, Qwen2.5 0.5B, Llama 3.2 1B, Llama 3.2 3B
Pick a smaller model for speed or a larger one for stronger answers.

Smaller models like TinyLlama 1.1B load fastest; larger ones like Llama 3.2 3B or Phi-3 mini give stronger answers but take a bigger first download. Switch models at any time.

ModelApprox. first download
TinyLlama 1.1Babout 0.7 GB
Qwen2.5 0.5Babout 0.5 GB
Llama 3.2 1Babout 0.9 GB
Llama 3.2 3Babout 2.3 GB
Phi-3 mini 3.8Babout 2.4 GB

Key features

  • Streaming replies: answers stream in token by token, generated on your own hardware.
  • In-session history: the chat history stays in this tab for the session and is not saved or synced anywhere.

← Back to utility tools

Related tools:

Tags: #utility

Related guides:

Loading reviews...

Frequently Asked Questions

Does my chat leave my device?

No. The model runs in your browser and your messages stay on your device.

Why is there a large download the first time?

The AI model weights (hundreds of MB) download once from the model registry and are then cached by your browser, so later visits start faster. The model is not bundled with the page.

Which browsers work?

Any browser with WebGPU - the latest Chrome or Edge on desktop, or Chrome on Android. Browsers without WebGPU show a notice instead; there is no CPU fallback.

Which model should I pick?

TinyLlama 1.1B is the smallest and fastest to load; Llama 3.2 3B and Phi-3 mini answer better but download more. You can switch models at any time.