GPT Image 2.5
Studio's everyday picture model — pictures from words or from many of your own pictures, up to the largest size, in the quality you pick.
GPT Image 2.5 is Studio's everyday picture model: pictures from your words, or from many of your own pictures with words, in the size and quality you choose. It is where Pictures begins on Studio's home, and the first of Images in Choose a model.
What it takes and makes
Make — words, and pictures if you like, to a picture
| Setting | What it takes |
|---|---|
| Your words | Required — up to 5,000 characters |
| Pictures | Up to 16 pictures (Image 1, Image 2… in the order you add them) |
| Size | 1K, 2K or 4K — starts at 1K |
| Quality | Low, Medium or High — starts at Low |
| Shape | Auto, 1:1, 3:4, 4:3, 2:3, 3:2, 9:16, 16:9 or 21:9 — starts at Auto |
| Naming your references | Type @ and pick one — one of each kind, read as “the picture”, “the clip” or “the sound” |
| Makes | A picture |
The price is never written here — prices change. It is on the model's page in Studio: on the Make button and beside each length, size or quality that changes it, before you press.
Writing for it
- Say what changes and what must stay — not everything twice. Your words have a limit; spend them on what matters.
- Words in the picture only when you ask. If you want lettering, quote the exact words, keep them short, and keep them off the subject that matters most.
- Shape starts at Auto — the model picks the frame; choose a shape when you need one.
- Quality starts at Low; the price follows the quality you pick, shown before you press.