Generate high-quality captions for your LoRA training datasets using AI. DatasetCaption connects to OpenRouter to have an LLM analyze and write detailed captions for your photos and videos — perfect for training LoRA models on state-of-the-art open source models like SDXL, Flux.2, Wan 2.2, and Z-Image Turbo. Supports text-to-image and image-to-image captioning modes, batch processing, video captioning, and fully customizable prompts.
Signed out
Sign in is recommended. Manual API key entry is still available below.
Select a model...
Selecting a preset fills the System Prompt. Saving stores it in your browser.
Advanced Options
Your API key is only used locally in this browser tab.
Idle