Type a rough idea, sentence, or brief and we generate a clean prompt with subject, scene, style signals, lighting, composition, and mood. Export plain text or structured JSON, then launch image or video generation from the same workbench.
當你手上還沒有參考圖，但又不想只停留在一句模糊想法時，就用這個工具。
結構化輸出可以把主體、場景、構圖、光線和情緒拆開，方便你在真正測試前先調整方向。
Image to Prompt needs a reference image. Text to Prompt is for moments when all you have is a sentence — a vibe, a brief, a creative direction — and you need a prompt that's actually usable in Midjourney, Flux, or Stable Diffusion.
Every generation returns a structured view (subject, scene, style, lighting, composition, mood) and a plain prompt. Tweak only the part you want to change instead of rewriting the entire prompt.
When the prompt is right, send it straight to text-to-image, image-to-image, text-to-video, or image-to-video without leaving the workbench.
Switch between Midjourney narrative style, Flux natural language, Stable Diffusion dense attribute lists, and JSON output depending on what your downstream model prefers.
Plain language is fine. Mention the subject, the scene, and any style direction you already have in mind.
Choose Midjourney, Flux, Stable Diffusion, or general — the wording adapts to what each model responds to best.
Open the structured prompt to fine-tune lighting, composition, or mood. Send the final prompt straight into image or video generation.
Text to Prompt is a free AI prompt generator that converts a rough idea or sentence into a structured prompt for Midjourney, Flux, Stable Diffusion, and Sora.
Image to Prompt extracts a prompt from a reference image. Text to Prompt builds a prompt from scratch using your written idea — useful when you don't have a reference image yet.
Yes. The same idea can be regenerated as a Midjourney narrative prompt, a Flux natural-language prompt, a Stable Diffusion dense prompt, or JSON without retyping your idea.
Yes. New accounts get free credits and the tool is included in every paid plan.