JIUFENG
Sign Up Free
Oopenai

GPT Image 2

GPT Image 2 is OpenAI's image generation model. In the JIUFENG workbench, create images from a text prompt, optionally with reference images, and choose an aspect ratio and 1K / 2K / 4K output.

Prompt
Reference images
Ratio and output size
1K0.1 Credits
2K0.3 Credits
4K1 Credits
Choose an output size before generating.

Visual inspiration: explore subjects, lighting and composition.

Woman in a beige sweater on a beach, with sunset light and waves behind her
Beach portrait: warm side lighting and a softly focused background.
Sunbeams passing through tall trees above a fern-lined forest path
Forest atmosphere: light beams, mist and layers of green foliage.
Child in an orange shirt holding a bubble wand on a sunny lawn
Outdoor portrait: a moving subject, bubbles and a blurred green background.
Snow-covered mountains reflected in a blue lake under golden light
Landscape composition: mountain reflections framed by trees and the shoreline.

Model Overview

GPT Image 2 is OpenAI's image generation model, and the first one JIUFENG integrated. Describe the image you want in words, and optionally attach reference images so the model can follow the subject, style or composition in them. Everything runs in the browser — nothing to install.

It is priced by output size: the 1K / 2K / 4K tiers cost different amounts, and larger costs more. That is the opposite of GPT Image 2.5, also on the site, where all three tiers cost the same. So while you are still exploring and iterating, the lowest tier here is the cheapest way to work; once the composition and prompt are settled, decide whether to move up a tier or switch models.

Workspace Options

Prompt

Describe the subject, setting, lighting and style in words.

Reference images

Attach reference images to be used alongside the prompt as input for this generation.

Ratio and output size

Choose an aspect ratio and output tier before submitting; the tier sets the rate for this generation.

What it suits

Cheap iteration

The lowest tier is the least expensive way to generate on the site. Iterate there while the direction is unsettled, then move up once the composition and prompt are fixed.

Reference images

Attach reference images to pin down a subject, a style or a composition, used together with the prompt.

Aspect ratio

Pick an aspect ratio before submitting, or let the model infer one from your reference image.

How to generate an image

  1. Open the workbench with this model selected and describe the subject, setting, lighting and style. Add reference images when you want to steer the result.
  2. Choose an aspect ratio and output tier — this model is priced per tier, so the tier sets what the generation costs. Check the estimate before you submit.
  3. Sign in and submit. Review the result in the workbench, then adjust the prompt or reference images for the next run.

Run in WorkspaceCompare image and video pricing

Getting good results

  1. The output tier sets the rate for the generation. Explore at a lower tier, then regenerate with the same prompt and reference images once the direction is settled.
  2. Naming the subject, setting, lighting and style explicitly works better than stacking adjectives.
  3. Choose reference images that actually convey the subject's features or the target style; you do not need to use the maximum number.

Specifications

Input
A text prompt, plus optional reference images — up to 15 per generation, 20MB each, as PNG / JPG / WebP, or HEIC (converted in your browser before upload).
Aspect ratio
15 presets, or follow the aspect ratio of your reference image.
Output
1K / 2K / 4K, chosen before you generate; rates at the top of the page.
Failure refund
Credits spent on a generation that fails because of a system error are returned automatically.

Use responsibly

Use JIUFENG in accordance with the Terms of Service and Acceptable Use Policy. Requests that do not meet applicable policies may be declined.