Skip to main content

​
Playground

The LLMTune playground lets you compare models side-by-side, test prompts, and share results.

​
Features

  • Model selection: Choose one or multiple models, including deployed fine-tunes.
  • Multi-modal support: Attach images for vision-capable models.
  • Attachments: Paste or upload images. LLMTune validates size and format.
  • Temperature controls: Adjust sampling parameters per session.
  • Transcript history: Export conversation logs to re-run through the API.

​
Workflow

  1. Open Playground from the dashboard.
  2. Select models to compare using the selector.
  3. Enter your prompt or conversation.
  4. Optionally attach images for multimodal evaluation.
  5. Send and observe responses side-by-side with latency data.
  6. Use Clear to reset the session or Copy to send results to teammates.

​
Clipboard and Upload Support

  • Drag-and-drop images into the composer.
  • Paste from clipboard (PNG, JPEG, WebP, GIF).
  • Remove attachments using the inline control next to each file.