Playground+
System prompt optional
Retrieval context optional
Prompt
Ctrl+Enter to run
Run a prompt to see token confidence
Each word in the output is colored by how confident the model was when choosing it. Green = near-certain, red = multiple alternatives were close.
Click any token to inspect the top alternatives and run parameters.
💡 Works with OpenAI, Ollama, LM Studio, or any OpenAI-compatible endpoint 💡 No API key needed for local models — use ⚙ Endpoint above