GPT Image 1.5

What it can do

• Creates a picture from a description.
• Edits a finished picture by text.
• Takes references: up to 16.
• Up to 4 variants per run.
• Quality: standard · HD.
• Formats: 1:1 · 2:3 · 3:2.

Prices

Price per one 1024×1024 image, calculated from the Studio price list on October 16, 2026: 1 ⚡ ≈ $0.025. The exact amount is shown on the Generate button before you start.

What it is good for

• Edits by text: swap a background, clothing or details without redrawing the whole frame.
• Frames with your character or object from references.
• Start frames for video: picture first, then animation.
• Cheap tests and drafts.

How to pick it in the Studio

1. Open the Studio

   Go to gibrid.ai/studio and log in.

2. Choose Photo

   The Photo / Video switch is at the bottom of the workspace.

3. Open the model list

   Click the current model name in the line above the input field and find "GPT Image 1.5".

4. Describe the frame

   Write one or two sentences and choose the format and quality in the settings.

5. Check the price and start

   The price in ⚡ is shown on the Generate button before you start.

Prompt tips

• Order: background and scene, then subject, key details, constraints, and the use case ("for an ad", "UI mock", "infographic").
• Pick one writing style, paragraphs or short items, and do not mix them.
• Refine with single-detail changes.

Common mistakes

• Recognisable real people and explicit or deceptive political scenes are blocked.
• Words like "stock photo" and "clip art" give a generic result.
• GPT Image 2 has no transparent background.

FAQ