Web, Data and AI Tools
Home
Analytics
◐
WebGPU LLM Chat
☆
Chat with AI models running entirely in your browser using WebGPU. No API keys, no server.
Your browser does not support WebGPU. Please use a recent version of Chrome, Edge, or another Chromium-based browser.
Model
SmolLM2-360M-Instruct (tiny, fast download)
SmolLM2-1.7B-Instruct (small, good quality)
Phi-3.5-mini-instruct (3.8B, high quality)
Llama-3.2-1B-Instruct (good balance)
Qwen2.5-1.5B-Instruct
Gemma-2-2B-Instruct (Google, 2B)
DeepSeek-R1-Distill-Qwen-7B (reasoning, ~5GB)
Load Model
Clear Chat
Initializing...
System Prompt
You are a helpful assistant.
Load a model to start chatting.
Send
Stop