Everything you need to get reliable outputs from both AI image tools.
Use this documentation to move from first upload to consistent results. It covers prompt generation, image description, mode selection, and practical quality improvements for real creative work.
Image to Prompt is best when you need a generation-ready prompt. AI Image Describer is best when you need analysis, extraction, or a direct answer about the image.
2. Pick the correct mode
Use general or structured modes for broad prompt drafting, Midjourney or Stable Diffusion modes for platform-specific phrasing, and object/text/art-style modes for targeted analysis.
3. Refine for your workflow
Treat the first result as a strong draft. Add constraints, remove noise, and adjust wording for your model, brand, or visual direction.
Prompt modes
General: a solid all-purpose prompt draft.
Structured: labeled sections for easier editing and reuse.
Graphic: more design-system and layout oriented outputs.
JSON: machine-readable structure for downstream workflows.