LocalSphere // Browser-Based AI Chatbot

1. Model Engine

Model Selection

Device Accelerator

2. Hyperparameters

Temperature 0.7

Max New Tokens 256

Top-P (Nucleus) 0.9

3. Bot Persona Preset

4. Engine Telemetry

Load Latency: -- sec

Generation Speed: -- tok/s

Session Tokens: 0

🔮

Welcome to LocalSphere

Your private browser-based chatbot. Select your LLM engine, choose WebGPU or WASM, and click a starter prompt below to run local models without any API keys.