# Flux 2 Pro

Black Forest Labs' FLUX.2 Pro: photoreal product scenes, exact hex colours and multi-reference composites.

Last updated: 2026-10-05

**Key facts**

- Flux 2 Pro is Black Forest Labs' FLUX.2 [pro], a production model for generating and editing images, strong on photorealism and fine detail.
- Flux 2 Pro makes 1K images in Sloosh and takes up to 8 reference images.
- Flux 2 Pro has a seed in Sloosh, so you can repeat a result, and takes prompts up to 32,768 characters.
- Flux 2 Pro takes exact colours as hex codes, each tied to an object, for brand-accurate product scenes.

Web page: https://sloosh.ai/docs/flux-2-pro

Flux 2 Pro is Black Forest Labs' production model for generating and editing. It is strong on photorealism and fine detail, takes exact colours as hex codes, and combines several reference images into one scene.

## When to use it

- Photoreal product scenes that must match exact brand colours.
- Combining a product, a setting and a style from separate references.
- Not for negative prompts: it doesn't support them. Describe what you want instead.

## How to prompt it

- It pays most attention to what comes first. Order the prompt: subject, action, style, context, then secondary details.
- 30 to 80 words is usually right; go longer only for a complex scene.
- Tie every hex colour to an object, for example "a towel in #9CAF88 sage green".
- Name references by number and job: "the bottle from image 1", "the counter in image 2". Put the base image first.
- For photorealism, name the camera, lens and aperture.
- Say "sharp focus throughout" rather than "no blur".

## Example prompt

With a product photo as image 1 and an empty counter as image 2:

**Prompt**

```text
Place the serum bottle from image 1 on the marble counter in image 2, label facing the camera, next to a folded hand towel in #9CAF88 sage green. Soft morning window light from the left, gentle shadows, shallow depth of field. Shot on Hasselblad X2D, 80mm lens, f/2.8.
```

## Settings in Sloosh

- Aspect ratio: 1:1, 2:3, 3:2, 16:9, 9:16, 4:3, 3:4, 4:5, 5:4 or 21:9.
- Resolution: 1K.
- References: up to 8 images.
- Seed: available, to repeat a result.
- Prompt: up to 32,768 characters.

Open it in [Image](https://sloosh.ai/studio/image) and pick Flux 2 Pro in the model chip. The Generate button shows the cost first.

## Frequently asked questions

### What is Flux 2 Pro best for?

Flux 2 Pro is best for photoreal product scenes that must match exact brand colours, and for combining a product, a setting and a style from separate reference images into one shot. Black Forest Labs positions it as production-grade image editing and generation.

### How do I match exact brand colours with Flux 2 Pro?

Give each colour as a hex code tied to one object, for example "a towel in #9CAF88 sage green" or "the bottle cap in #0047AB". A hex code with no object to attach it to is a common Flux 2 Pro mistake, because the model can't tell where the colour belongs.

### How long should a Flux 2 Pro prompt be?

Black Forest Labs says 30 to 80 words is usually right for Flux 2 Pro; go longer only for a complex scene. Flux 2 Pro pays most attention to what comes first, so lead with the subject and action, then style and context, and leave secondary details to the end.

### How do I combine several reference images with Flux 2 Pro?

Name each reference by number and job, such as "the bottle from image 1" and "the counter in image 2", and say where things go and how they interact. Put the base image first. Flux 2 Pro takes up to 8 references in Sloosh.

### How do I get photoreal results from Flux 2 Pro?

Name the camera, lens and aperture, for example "Shot on Hasselblad X2D, 80mm lens, f/2.8", or a film stock such as Kodak Portra 400. Describe the light by direction and softness. Flux 2 Pro has no negative prompt, so write "sharp focus throughout" rather than "no blur".

## Sources

- [BFL: FLUX.2 prompting guide](https://docs.bfl.ml/guides/prompting_guide_flux2)
- [BFL: multi-reference editing](https://docs.bfl.ml/guides/prompting_editing_multi_reference)
- [BFL: FLUX.2 overview](https://docs.bfl.ml/flux_2/flux2_overview)
