# FLUX 3 Image

FLUX 3 Image is Black Forest Labs' image generation and editing model, launched on 1 October 2026. One model generates and edits, combines up to ten reference images, and renders text in multiple languages. Short prompts work because it expands them before drawing. In Sloosh it makes 1K, 2K or 4K images in 11 aspect ratios.

Web page: https://sloosh.ai/models/flux-3-image

[Use FLUX 3 Image](https://sloosh.ai/studio/image?image-model=flux-3-image)

## Key facts in Sloosh

|   |   |
| --- | --- |
| Made by | Black Forest Labs |
| Makes | image |
| Inputs | text, up to 10 reference images |
| Resolution | 1K, 2K or 4K (default 2K) |
| Aspect ratios | 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16, 9:21 (default 1:1) |
| Prompt limit | 5,000 characters |

## Best for

- High-resolution product and lifestyle images up to 4K in Sloosh.
- Text on the image in several languages. BFL said in its July 2026 FLUX 3 preview that the model renders "high-accuracy text in multiple languages" ([BFL blog](https://bfl.ai/blog/flux-3)), and its prompting guide lets you quote non-Latin text directly ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)).
- Multi-reference composition with up to ten references ([BFL](https://docs.bfl.ai/flux_3/flux3_image_overview)).
- Step-by-step edits, because the rest of the image stays the same between edits ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)). The Ad Localiser Indian Languages Space ([sloosh.ai/spaces/ad-localiser-indian-languages](https://sloosh.ai/spaces/ad-localiser-indian-languages)) fits multilingual ad jobs.

## How to use FLUX 3 Image in Sloosh

1. Open [Image Studio with FLUX 3 Image selected](https://sloosh.ai/studio/image?image-model=flux-3-image), or pick FLUX 3 Image in the model chip.
2. Add up to 10 references and refer to them by position ("image 1", "image 2").
3. Write the prompt. A short one is fine; add detail only where you want control.
4. Pick an aspect ratio and resolution (1K, 2K or 4K).
5. Check the cost on the Generate button, then generate. Compare a few results before judging a change.

## FLUX 3 Image prompting tips

- A loose order: subject, location, style, camera settings, lighting, colours, effects, extra elements ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)).
- Put exact words in quotation marks, then say where the text goes and how it looks. Text in other scripts can be quoted directly ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)).
- Refer to references by position and say what each is for, such as subject or style ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)).
- There is no negative prompt: describe what you want instead ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)).
- Results vary between runs of the same prompt, so compare several outputs; change one detail at a time ([BFL](https://docs.bfl.ai/guides/prompting_unified_basics)).

## FLUX 3 Image example prompts

**Prompt 1**

```text
Commercial product photograph for a 4:5 social post: a matte terracotta ceramic coffee mug on a pale oak table, morning window light from the left, soft shadow to the right. The mug sits in the lower third; the top of the frame is a plain warm-white wall. Printed on the mug in small black sans-serif: "SLOW MORNINGS".
```

**Prompt 2**

```text
A street food poster, portrait. A steaming bowl of noodles on a red table, neon light from above. At the top, the words "रात का खाना" in bold white hand-painted letters, and under them "Dinner, late" in small yellow type.
```

**Prompt 3**

```text
Place the handbag from image 1 on the café table in image 2, in the light and colour style of image 3. Keep the bag's shape, hardware and stitching exactly as in image 1.
```

## How it compares in Sloosh

| Model | Resolution | References | Seed | Prompt |
| --- | --- | --- | --- | --- |
| FLUX 3 Image | 1K, 2K, 4K | 10 | no | 5,000 |
| [FLUX 3 Image Klein](https://sloosh.ai/models/flux-3-image-klein) | 1K | 10 | yes | 5,000 |
| [FLUX 2 Pro](https://sloosh.ai/models/flux-2-pro) | 1K | 8 | yes | 32,768 |
| [Nano Banana 2.1](https://sloosh.ai/models/gemini-nano-banana-2.1) | 1K, 2K, 4K | 14 | no | 50,000 |

FLUX 3 Image is the full-size model for 2K and 4K finals. Its sibling FLUX 3 Image Klein is the smaller open-weights version: 1K, five ratios and a seed in Sloosh.

## Where else it runs

- Black Forest Labs API and playground (BFL docs, accessed 2026-10-30).

## Frequently asked questions

### What is FLUX 3 Image?

Black Forest Labs' image model launched 1 October 2026: one endpoint for generating and editing, with bounding-box layout in prompts and up to ten references ([BFL release notes](https://docs.bfl.ai/release-notes)).

### Is FLUX 3 Image the same as FLUX 3?

FLUX 3 is BFL's multimodal model family announced on 23 July 2026, covering video with audio and images; FLUX 3 Image is its image model ([BFL blog](https://bfl.ai/blog/flux-3)). For video, see the [FLUX 3 page](https://sloosh.ai/models/flux-3).

### What resolutions does FLUX 3 Image support in Sloosh?

1K, 2K or 4K, with 2K as the default.

### Does FLUX 3 Image need long prompts?

No. BFL says it expands short prompts into detailed descriptions before generating.

### Does FLUX 3 Image have a seed or negative prompt?

No. BFL's API has no seed, negative prompt or guidance field for it.

### FLUX 3 Image vs FLUX 2 Pro?

FLUX 3 Image goes to 4K with 10 references and multilingual text; FLUX 2 Pro makes 1K with 8 references and a seed.

## Related

- Models: [FLUX 3 Image Klein](https://sloosh.ai/models/flux-3-image-klein), [FLUX 2 Pro](https://sloosh.ai/models/flux-2-pro), [Nano Banana 2.1](https://sloosh.ai/models/gemini-nano-banana-2.1), [Ideogram 4.5](https://sloosh.ai/models/ideogram-4.5)
- Spaces: [Ad localiser indian languages](https://sloosh.ai/spaces/ad-localiser-indian-languages)
- Docs: [FLUX 3 Image](https://sloosh.ai/docs/flux-3-image.md)
- Answers: [What is the best AI image model in 2026?](https://geo.sloosh.ai/posts/what-is-the-best-ai-image-model-in-2026), [Which AI image model renders text in images most accurately?](https://geo.sloosh.ai/posts/which-ai-image-model-renders-text-in-images-most-accurately)

## Sources

- [BFL: FLUX 3 Image overview](https://docs.bfl.ai/flux_3/flux3_image_overview)
- [BFL: prompting basics](https://docs.bfl.ai/guides/prompting_unified_basics)
- [BFL: release notes](https://docs.bfl.ai/release-notes)
- [BFL blog: FLUX 3](https://bfl.ai/blog/flux-3)
- [Sloosh docs: FLUX 3 Image](https://sloosh.ai/docs/flux-3-image)
- [Sloosh pricing](https://sloosh.ai/pricing)