LLM Playground

Run WebLLM models directly in your browser — completely private, no server calls.

⚠️ First Time Setup: Model download may take 5-10 minutes depending on your internet speed and device. This is a one-time setup. Models are cached in your browser's local storage.

Select Model

Llama 2

7B • ~4 GB

Mistral

7B • ~4 GB

RedPajama

3B • ~2 GB

Generation Parameters

Temperature

0.70 (randomness)

Top P

0.90 (diversity)

Max Tokens

512 tokens

System Prompt

Conversation

WebLLM Playground: Select a model and start chatting. All processing happens locally in your browser. Adjust temperature for creativity (0.0 = deterministic, 1.0 = random) and top-p for diversity.