# GPT Image 2.5 Flare

OpenAI's fast GPT Image 2.5 tier for everyday images and edits, including layouts with text.

Last updated: 2026-10-05

**Key facts**

- GPT Image 2.5 Flare is the fast tier of OpenAI's GPT Image 2.5; OpenAI says it is up to 50% faster than GPT Image 2 at comparable quality.
- GPT Image 2.5 Flare takes up to 16 reference images in Sloosh Studio and prompts up to 32,000 characters.
- GPT Image 2.5 Flare offers resolution 1K, 2K or 4K and quality low, medium, high, xhigh or max in Sloosh, the same as GPT Image 2.5 Sunburst.
- GPT Image 2.5 Flare offers aspect ratios Auto, 1:1, 3:2, 2:3, 16:9, 9:16, 4:5 and 5:4 in Sloosh.

Web page: https://sloosh.ai/docs/gpt-image-2-5-flare

GPT Image 2.5 Flare is the fast tier of OpenAI's GPT Image 2.5. OpenAI says it is up to 50% faster than GPT Image 2 at comparable quality. It takes the same prompt and settings as Sunburst.

## When to use it

- Everyday images, quick drafts and trying several directions.
- Layouts with a headline or a short line of copy, such as banners and newsletter headers.
- Move to [GPT Image 2.5 Sunburst](https://sloosh.ai/docs/gpt-image-2-5-sunburst.md) when fine detail has to survive a close look.

## How to prompt it

- Describe the scene, then the subject and details, then any text and where it goes.
- Put literal text in quotes. For an unusual word, spell it out letter by letter.
- Leave room in the layout for text by saying where the empty space is, as in the example.
- Name references by number and job ("Image 1: product photo"), and say what must not change.

[Prompting GPT Image](https://sloosh.ai/docs/prompting-gpt-image.md) has the full guide.

## Example

Text only: a flat lay with a headline placed in the empty space.

![A flat lay of a straw hat, sunglasses, a linen shirt, a book and sunscreen on sand-coloured paper, with the words SUMMER EDIT and New arrivals every Friday on the right](https://sloosh.ai/docs/m-gpt-image-2-5-flare.webp)

_16:9, quality auto, Enhance off._

**Prompt**

```text
A bright flat-lay photo for a summer newsletter header. On a pale sand-coloured background: a straw hat, round sunglasses, a linen shirt folded neatly, a paperback book and a glass bottle of sunscreen. Hard midday sunlight with crisp shadows. In the empty space on the right, the words "SUMMER EDIT" in bold black sans-serif capitals, with "New arrivals every Friday" in small text underneath.
```

## Settings in Sloosh

- Aspect ratio: Auto, 1:1, 3:2, 2:3, 16:9, 9:16, 4:5 or 5:4.
- Resolution: 1K, 2K or 4K.
- Quality: low, medium, high, xhigh or max.
- At 2K or 4K, an Auto aspect ratio renders square.
- References: up to 16 images.
- Prompt: up to 32,000 characters.

Open it in [Image](https://sloosh.ai/studio/image) and pick GPT Image 2.5 Flare in the model chip. The Generate button shows the cost first.

## Frequently asked questions

### What is GPT Image 2.5 Flare best for?

GPT Image 2.5 Flare is best for everyday images, quick drafts and trying several directions, including layouts with a headline or a short line of copy such as banners and newsletter headers. When fine detail has to survive a close look, rerun the prompt on [GPT Image 2.5 Sunburst](https://sloosh.ai/docs/gpt-image-2-5-sunburst.md).

### How do I put text in an image with GPT Image 2.5 Flare?

Put the exact words in quotes, say where they go and describe the type, such as bold black sans-serif capitals. Tell GPT Image 2.5 Flare where the empty space is so the text has room. Spell an unusual brand name letter by letter, and say the text appears once.

### Can GPT Image 2.5 Flare edit an existing photo?

Yes. Add the photo as a reference, name it by number and job ("Image 1: product photo"), describe the change and list what must not change. GPT Image 2.5 Flare takes up to 16 references in Sloosh Studio, so you can also add a style or setting image as Image 2.

### How do I combine several reference images with GPT Image 2.5 Flare?

Call each reference by its position and give it a job: "the bag in image 1", "the model in image 2", "the light in image 3". Then say what goes where and what comes from which image. In Studio, GPT Image 2.5 Flare reads references in the order you added them.

## Sources

- [OpenAI image prompting guide](https://developers.openai.com/api/docs/guides/image-prompting)
- [fal: how to use GPT Image 2.5](https://fal.ai/learn/tools/how-to-use-gpt-image-2-5)
