Does anything I type get sent to a server?
No. The model runs on your GPU inside the browser. The only network traffic is the one-time download of model files from Hugging Face and the model library from GitHub.
Which browsers work?
Any browser with WebGPU: current Chrome and Edge on desktop and Android, Safari on recent macOS and iOS, and Firefox versions that ship WebGPU. If something is missing, the tool tells you what.
How much space and memory do I need?
The smallest models are about 200–400 MB and need under 2 GB of GPU memory. Mid-size models are 1–2.5 GB and suit 8 GB machines. The catalog shows the exact numbers for every model and whether it fits your device.
Can I use it offline?
Yes. Once a model is downloaded, the browser keeps it and the chat works without a connection.
How do I add my own model?
Open Models, choose Add model, and paste the URL of a repository with MLC-compiled weights (mlc-chat-config.json plus tensor-cache.json) and the WebGPU model library built for it. The tool checks the files before it saves anything.
What is the Reasoning block?
Some models (Qwen3, Qwen3.5, DeepSeek R1) think before they answer. You see that thinking in a collapsible block above the answer, and it is never sent back to the model.
Why is a model marked as not fitting?
The tool compares the model's GPU memory need with your device, checks for GPU features such as 16-bit shaders, and looks at free storage. You can still try it, and the tool will tell you if the GPU runs out of memory.
Where are my chats stored, and how do I delete them?
In this browser's IndexedDB. Delete a chat from its menu, or everything at once from Settings. Clearing the site's data in your browser removes them too.
Can I stop a long answer or cancel a download?
Yes. Stop replaces the send button while an answer streams, and every download has a Cancel button. Finished parts stay cached, so a retry picks up where it left off.
