ImgX - Image to Prompt, From Inspiration to Creation
https://imgxapp.com/ · Art & Design
Media
5 items
Description
Image to Prompt for AI art — turn any reference image into editable bilingual prompts for Midjourney, Stable Diffusion, and more. ImgX is an Image to Prompt extension and AI prompt generator for digital creators. Turn any reference image into editable Chinese and English AI art prompts for Midjourney, Stable Diffusion, and other image-generation tools. It analyzes the subject, pose, composition, style, lighting, colors, materials, and other visual details. Then use AI Prompt Rewrite to refine the current prompt with natural-language instructions—for example, change a pose, outfit color, or product detail. ImgX creates prompts; it does not generate images. Features 🖱 Hover over an image, capture a region, or upload a reference. ☁ Built-in Cloud mode is the default — sign in for free credits; no API key required. ✨ AI Prompt Rewrite — refine a prompt with natural-language instructions (Cloud only; 1 credit per submission). 🧠 Structured analysis — visual fields, bilingual prompts, and one-click copy. 📋 Batch queue — process images in parallel with automatic retries. 🔑 Optional BYOK — use your own OpenAI-compatible Vision API endpoint, key, and model for image analysis. Prompt Rewrite requires Cloud mode. Privacy Cloud is the default for new users. After sign-in, ImgX services process your email, credit records, and image-analysis and rewrite requests. Image pixels are used to process a request and are released when processing ends; a small thumbnail is stored only if you explicitly save a favorite. Payment credentials are handled by the payment provider; ImgX keeps only the minimum order records needed for credit settlement. Optional product analytics require your consent. No ads or cross-site tracking. In optional BYOK mode, image-analysis requests go directly from your browser to your chosen provider, and your API key stays on your device. See the privacy policy. Needs For Cloud, sign in with an email. Optional BYOK requires an OpenAI-compatible Vision API endpoint, API key, and model.