GPT Image

OpenAI's image models from ChatGPT: GPT Image 2 for layouts, exact text and faithful edits, with 1.5, 1 and Mini for cheaper runs.

Whisperer's verdict: 4.4 / 5

What it nails

• Follows instructions to the letter
• Clean logos and exact text
• Edits that change one thing only

Watch out for

• Top quality costs more per picture

GPT Image is OpenAI's image family, the engine behind images in ChatGPT. Where other models shine at mood, GPT Image shines at following instructions to the letter: exact words on a billboard, a clean logo, one colour changed and nothing else touched.

The Lab runs GPT Image 2 (the current flagship, released 21 April 2026), 1.5, 1 and 1 Mini, and you pay per picture.

What GPT Image is secretly great at

Diagrams you can actually learn from

OpenAI built GPT Image 2 for "stronger editing, better layouts, improved text rendering", including diagrams and infographics. Ask it to explain how something works and it draws the steps and labels them.

Create a detailed Infographic of the functioning and flow of an automatic coffee machine like a Jura. From bean basket, to grinding, to scale, water tank, boiler, etc. I'd like to understand technically and visually the flow.

Example by OpenAI · source ↗

Translate a finished design

Every label changes language while the drawing stays put. The secret is the last sentence: "Do not change any other aspect of the image."

Translate the text in the infographic to Spanish. Do not change any other aspect of the image.

Example by OpenAI · source ↗ · This is an edit: add your own infographic as a reference picture first.

Words that come out exactly right

Write the text in quotes, say "EXACT, verbatim", and describe the typography. OpenAI's own guide also suggests High quality for dense text.

Create a realistic billboard mockup of the shampoo on a highway scene during sunset. Billboard text (EXACT, verbatim, no extra characters): "Fresh and clean" Typography: bold sans-serif, high contrast, centered, clean kerning. Ensure text appears once and is perfectly legible. No watermarks, no logos.

Example by OpenAI · source ↗ · This is an edit: add your own product photo as a reference picture first.

Logos that read at any size

Ask for vector-like shapes, a strong silhouette and "simplicity over detail", and GPT Image draws a clean mark instead of a busy illustration. (The model can also return a transparent background; in the Lab your logo comes on a plain background for now.)

Create an original, non-infringing logo for a company called Field & Flour, a local bakery. The logo should feel warm, simple, and timeless. Use clean, vector-like shapes, a strong silhouette, and balanced negative space. Favor simplicity over detail so it reads clearly at small and large sizes. Flat design, minimal strokes, no gradients unless essential. Fully transparent background. Deliver a single centered logo with generous padding, clean alpha edges, and no solid backdrop, scenery, checkerboard, or watermark.

Example by OpenAI · source ↗

Edits that change one thing only

Name the change, then repeat "do not change anything else". Faces, light and composition stay the same.

Change the color of the red hat to light blue as velvet. Do not change anything else.

Example by OpenAI · source ↗ · This is an edit: add your own photo as a reference picture first.

Which GPT Image to pick

How to prompt GPT Image

From OpenAI's official prompting guide.

1. Use a fixed order: scene, then subject, then key details, then limits.
2. Name the medium and materials: photo, watercolour, 3D render; wool, brushed steel.
3. For realism, write "photorealistic" and describe the camera and light.
4. Put text in quotes or CAPITALS and describe the font.
5. List what must not change in a separate sentence ("do not change anything else").
6. With several pictures, number them: "image 1", "image 2".

Photos that feel honest

Ask for texture and imperfection. OpenAI's example says "no glamorization, no heavy retouching", and the result looks like a real 35mm shot.

Create a photorealistic candid photograph of an elderly sailor standing on a small fishing boat. He has weathered skin with visible wrinkles, pores, and sun texture, and a few faded traditional sailor tattoos on his arms. He is calmly adjusting a net while his dog sits nearby on the deck. Shot like a 35mm film photograph, medium close-up at eye level, using a 50mm lens. Soft coastal daylight, shallow depth of field, subtle film grain, natural color balance. The image should feel honest and unposed, with real skin texture, worn materials, and everyday detail. No glamorization, no heavy retouching.

Example by OpenAI · source ↗

Combine and restyle your own pictures

Upload the pictures, number them in the prompt, and say which one gives the scene and which gives the subject or the style.

Place the dog from the second image into the setting of image 1, right next to the woman, use the same style of lighting, composition and background. Do not change anything else.

Example by OpenAI · source ↗ · This is an edit: add your own two photos (the scene, then the dog) as a reference picture first.

Use the same style from the input image and generate a man riding a motorcycle on a white background.

Example by OpenAI · source ↗ · This is an edit: add your own style picture as a reference picture first.

FAQ