GPT Image 2 is an AI image generator and editor available within ImgEditor that creates, edits, and upscales images from text prompts and reference images directly in a browser.
What is GPT Image 2?
GPT Image 2 is a generative AI model accessible through the ImgEditor online workspace. It takes text descriptions, uploaded reference images (up to 8 per job), or existing photos as input and outputs edited or newly generated images at 1K or 2K resolution. The model supports pixel-accurate text rendering in 48+ languages, making it suitable for multilingual product labels, posters, and ads.
Key Features
- Text‑to‑Image generation — Describe a scene, product, or style in natural-language prompts (up to 5000 characters) and generate visuals at 1:1, portrait, or landscape aspect ratios.
- Image‑to‑Image editing — Upload an existing photo and modify backgrounds, lighting, colors, or objects with text commands; ImgEditor preserves shadows, reflections, and depth.
- Reference‑guided workflows — Use up to 8 product shots or environment references to guide composition, subject style, and layout while generating new images.
- Image upscaling — Enhance details and clarity of generated or uploaded images for sharper previews, product shots, and social media graphics.
- Integrated editing tools — Replace backgrounds, adjust lighting and color temperature, change product colors, and clean up product photos without leaving the workspace.
- Specialised generators — Packaging mockup, poster, accurate‑text image, and storyboard creation tools are built into the same workflow.
Who is it for?
- E‑commerce teams — Create product hero shots, lifestyle scenes, and ad visuals from reference images or prompts without a photo studio.
- Designers and marketers — Generate campaign posters, social graphics, and packaging concepts with brand‑consistent color palettes and readable text.
- Content creators — Produce thumbnails, parody images, or storyboards by uploading a selfie and describing a new setting or style.
What can you do with GPT Image 2?
- Product photography: Upload a single product shot and prompt ImgEditor to place it on a water mirror surface or in a natural stone setting with soft greenery.
- Image editing: Take an existing photo and change the background to a conference room, adjust lighting to golden hour, or recolor packaging to match a brand palette.
- Poster and ad creation: Generate a luxury beauty advertisement or a snack e‑commerce ad that preserves the original product shape, packaging, and logo.
How does GPT Image 2 work?
After creating a free ImgEditor account, users start with 30 credits (no credit card required). Drag up to 8 reference images into the workbench, type a natural‑language prompt describing the desired output, then click Generate. Results appear in seconds and can be refined iteratively. Exports are watermark‑free.
Pricing
ImgEditor offers a free tier with 30 starter credits. Paid yearly subscriptions include a 25% discount and additional credits for higher‑volume generation and editing.
FAQ
Is GPT Image 2 free to use?
New users receive 30 free credits upon registration — enough to generate roughly two standard images. No credit card is required for the free tier. Additional credits are available through paid plans.
What output resolutions are supported?
ImgEditor generates images at 1K and 2K resolution. 2K output is recommended for hero product shots, posters, and social assets that require extra detail.
Can I edit an image I already have?
Yes. Upload an existing photo into the editor, then describe changes — such as “replace the background with a beach at sunset” or “change the product color to deep blue” — and the model applies the edit while preserving realistic shadows and reflections.
