What is FLUX.2 [pro]?

It is the production-grade model in Black Forest Labs' second FLUX generation, released on 25 November 2025, built to balance quality and speed. Black Forest Labs sped it up roughly twofold in March 2026.

Can it write text into images?

Yes. Our poster test quoted a headline and a small line and both came back spelled exactly as written, in the typeface style the prompt described. Quote the words and say where they sit.

Does it follow colour codes?

Yes. Hex codes written in the prompt steer the palette. In our poster test a dusk teal and a copper horizon band came out in the tones we specified.

Can it make NSFW images?

Only up to a point. The Mild level renders and the Moderate level sometimes did, but Strong-level requests were toned down in our tests, and the Very strong level was refused. Editing refused portrait sources even for a Mild-level outfit change. Refused requests are not charged. For fewer restrictions, see the open-weight FLUX.2 [dev] and [klein] models, which refused nothing in our tests.

How many reference images can I use?

One to three in an edit. Say what each one contributes: image 1 is the person, image 2 the product, image 3 the place.

Does the aspect ratio change the price?

No. Text to image is billed per image by size tier, and the shape does not matter. An edit is a flat price per image whatever the number of sources or the size tier.

Does the seed repeat an image?

A fixed seed is for comparing prompts side by side; it does not promise an identical result.

Does it search the web or have adjustable steps?

No. Those options exist on other FLUX.2 models and in Black Forest Labs' own tooling, but are not offered here.

FLUX.2 Pro API: Photoreal Images With Clean Text

black-forest-labs/flux-2-pro/text-to-image

FLUX.2 [pro] text to image turns a written brief into a photoreal picture at 1k or 2k in nine aspect ratios, with readable in-image text and hex-code colour control.

From $0.03/image

Run model View API

Input

Text to image

Prompt596/5000

Aspect ratio

Resolutionaffects price

Seed

Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run

Example result

Animate

Edit

Split into layers

{
  "model": "black-forest-labs/flux-2-pro/text-to-image",
  "input": {
    "aspect_ratio": "3:2",
    "resolution": "1k",
    "seed": 8,
    "prompt": "Two women in their thirties share a slow morning on a narrow wrought-iron balcony above a Lisbon street with tiled facades. One, with long dark curls, wears an ivory satin slip dress with one thin strap slipped off her shoulder and leans on the railing holding a small espresso cup; the other, with a cropped platinum pixie cut, wears an oversized unbuttoned pale-blue linen shirt over a white tank top and rests her chin on her partner's shoulder from behind. Soft early sunlight, warm terracotta rooftops and laundry lines in the background, candid documentary photograph, natural skin texture."
  }
}

Introduction

FLUX.2 [pro] Text to image is Black Forest Labs' production model from the FLUX.2 generation, released on 25 November 2025. It turns a written prompt into a photographic image at 1k or 2k in nine aspect ratios, with text that reads correctly inside the picture and colours you can set with hex codes.

Choose a mode by starting material

What you have	Mode
a written brief	FLUX.2 [pro] Text to image
one to three images to change or combine	FLUX.2 [pro] Edit

Key features & highlights

Readable text in the picture

Quote the words and say where they sit. In our poster test a serif headline and a small sans-serif line both came back spelled exactly as written.

Hex-code colour control

Write colours as hex codes and the palette follows. Our poster used a teal and a copper tone, and the sea and the horizon band took them.

Documentary realism

Skin texture, window light and film grain read as photographs. Our balcony portrait kept freckles and fine lines at 2k.

Moderated, with limits

The Mild level renders. In our tests the Moderate level rendered in some runs and came back toned down in others, Strong-level requests came back softened, and the Very strong level was refused without charge.

Request parameters

Request parameters for FLUX.2 Pro · Text to Image
Parameter	Required	Description
`prompt`	Yes	Required, up to 5,000 characters. Describe subject, clothing, setting, light and lens in plain sentences. at least 1 characters · at most 5000 characters
`aspect_ratio`	No	One of nine shapes: 1:1 (default), 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9, 9:21. The shape does not change the price. `1:1` · `16:9` · `9:16` · `4:3` · `3:4` · `3:2` · `2:3` · `21:9` · `9:21` · Default: `1:1`
`resolution`	No	1k (default) is about one megapixel; 2k is 2048 pixels on the long side. The size tier affects the price. `1k` · `2k` · Default: `1k`
`seed`	No	Optional whole number from 0 to 2147483647. Reuse it while changing the prompt to compare variations side by side; it does not guarantee an identical image. `≥ 0` · `≤ 2147483647`

Pricing

Billed per image by size tier; the aspect ratio does not change the price. Refused and failed requests are not charged.

Pricing
Specification	Price
`1k`	$0.03 / image
`2k`	$0.0391 / image

Use cases

Posters and campaign drafts

A headline, a tagline and a photograph in one frame, for fragrance, fashion or product concepts.

Couples and character portraits

Two people with distinct hair, clothing and mood in a candid scene, for stories, mood boards and character sheets.

Social visuals in any shape

Square, vertical, wide and ultra-wide frames from the same prompt by changing the shape.

Explore key use-case tutorials

Prompt guide & pro tips

Write full sentences

Subject, clothing, setting, light and lens as sentences works better than a list of keywords.

Quote text and name its style

Put each line in quotes and describe the type, for example a wide-spaced serif headline at the top.

Pin the palette

Give hex codes for the two or three colours that matter and say where they appear.

Draft at 1k

Test a composition at 1k, then render the keeper at 2k. The size tier affects the price.

Use a seed to compare

Keep the seed fixed while you change one phrase, so differences come from your wording.

API quick start

# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $SPICY_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
  -d '{
  "model": "black-forest-labs/flux-2-pro/text-to-image",
  "input": {
    "prompt": "A ceramic teapot on a linen cloth, soft window light, muted palette",
    "resolution": "1k",
    "aspect_ratio": "1:1"
  }
}'

# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $SPICY_API_KEY"

See the API reference for authentication, the complete schema, polling, and error handling.

Missing the model you need?

Tell us what to add and where you would use it. Requests go to the people who decide what gets onboarded next.

BLACK FOREST LABS IMAGE MODEL

FLUX.2 Pro API at a glance

FLUX.2 [pro], Black Forest Labs' production workhorse for photoreal portraits, readable text and exact brand colours.

Reference images: 3; Combine up to three sources, such as a person, a product and a setting, in one edit.
Largest size tier: 2K; 1k is about one megapixel; 2k reaches 2048 pixels on the long side.
Aspect ratios: 9; From 21:9 to 9:21, including square, 4:3 and 3:2.
Colour control: Hex; Write colours as hex codes in the prompt and they are followed closely.

Results

What you can make with FLUX.2 Pro

Text and colour
A fragrance poster with correct type
A man waist-deep in a dusk sea, with a wide-spaced serif headline reading SALT & CEDAR and a small line reading EAU DE PARFUM. Both lines came back spelled as quoted, and the teal and copper hex codes set the palette.
Text to image, 1k, 2:3.
Portraits
Morning light, natural skin
Two women on a Lisbon balcony in early morning light. Freckles, fine lines and early sun on tiled facades read as a documentary photograph at 1k.
Text to image, 2k, 3:2.
Three-image composite
Person, product and place in one ad
A portrait, a pair of headphones and a night rooftop merged into one advertisement. His face, hair and cardigan and the headphones' sand and gold design stayed as in the sources, and the headline HEAR THE CITY was added on top.
Edit, three sources, 2k, 3:4.
Relighting
Noon street to rainy neon night
One photo of a woman on a sunny pavement became the same street at night with wet neon reflections, rain, and drops on her blazer, while her face, outfit and pose stayed the same.
Edit, one source, auto shape.

Content test

Is FLUX.2 Pro uncensored? What we tested

We ran each level on our live setup and checked every clip frame by frame. A pass means the content was actually on screen, not just that the request went through.

Content levels tested on FLUX.2 Pro, by mode
Level	Image editing	Text to image
MildThe lightest prompts in the test	RefusedTested Oct 8, 2026	Not tested yet
ModerateWhere filtered models start to refuse	RefusedTested Oct 8, 2026	Not tested yet
StrongThe middle of the scale	RefusedTested Oct 8, 2026	Not tested yet
Very strongWhere most models start to hold back	RefusedTested Oct 8, 2026	RefusedTested Oct 8, 2026
MaximumThe top of the scale	Not tested yet	Not tested yet

Renders: on screen as written. Softened: generated, but milder than what was written. Refused: rejected. Not tested yet: we have not run that level on that mode.

How each mode behaves

All modes: Black Forest Labs applies its own moderation to every prompt and result. In our tests text to image rendered the Mild level, rendered the Moderate level in some runs and toned it down in others, softened Strong-level requests, and refused the Very strong level. Refused requests are not charged.
Image editing: Editing is stricter than text to image. In our tests portrait sources were refused even for a Mild-level outfit change. Use editing for products, places, and portraits where you change only the background, light or style.
All modes: Most text-to-image requests finished in 20 to 90 seconds, and edits in 45 to 120 seconds, longer when the queue is busy. Output is JPEG. There is no negative prompt, no step or guidance setting, and no more than three sources per edit.

Use cases

What to build with the FLUX.2 Pro API

Posters and campaign visuals
Fragrance, fashion and product posters where a headline and a small line of type share the frame with a photograph.
Documentary-style portraits
Couples, friends and solo portraits with natural skin texture, window light and film grain.
Product and brand composites
Put a person, a product and a location from three separate photos into one advertisement.
Relighting and scene changes
Turn a daytime street into a rainy night, keeping the person, the outfit and the pose as they were.

Prompts

FLUX.2 Pro prompt examples you can copy

Paste one into the playground as it is, then change one thing at a time.

Text to image
A fragrance poster with correct type
A fragrance advertising poster. A man in his thirties with wet swept-back dark hair and light stubble stands waist-deep in a calm sea at dusk, wearing a white linen shirt fully unbuttoned and soaked through, looking past the camera. Muted teal sky and water in #1B3A4B with a thin band of copper light on the horizon in #C97B4A. Clean minimalist layout: an elegant wide-spaced serif headline at the top reads "SALT & CEDAR", and smaller sans-serif text at the bottom reads "EAU DE PARFUM". Photographic, high-end campaign look, correctly spelled text.
Why it works. A tall poster at 1k with both lines of type quoted and placed, and two hex codes for the palette. Both lines came back spelled as written and the sea and horizon took the specified tones.
Try this prompt
Text to image
Two women, one slow morning
Two women in their thirties share a slow morning on a narrow wrought-iron balcony above a Lisbon street with tiled facades. One, with long dark curls, wears an ivory satin slip dress with one thin strap slipped off her shoulder and leans on the railing holding a small espresso cup; the other, with a cropped platinum pixie cut, wears an oversized unbuttoned pale-blue linen shirt over a white tank top and rests her chin on her partner's shoulder from behind. Soft early sunlight, warm terracotta rooftops and laundry lines in the background, candid documentary photograph, natural skin texture.
Why it works. A wide 1k frame that describes each woman by hair, clothing and pose. The result keeps freckles and fine lines, with low sun behind the rooftops.
Try this prompt
Text to image
Kitchen dance at 2am
Candid 35mm film photograph of a woman and a man in their thirties dancing barefoot in a small apartment kitchen at 2am. She wears an oversized white men's dress shirt with rolled sleeves, hair falling loose; he wears grey sweatpants and a fitted white undershirt, one hand on her waist, foreheads almost touching, both laughing. A single warm pendant lamp, the open fridge spilling cool light across the floor, dishes in the sink, a record player on the counter. Soft grain, natural imperfect framing.
Why it works. Two light sources, a warm lamp and the cold fridge, give the frame its mood. The film grain and loose framing read as a snapshot rather than a studio shot.
Try this prompt
Image editing
Person, product and place in one ad
Create a headphone advertisement. The man from image 1 wears the headphones from image 2 over his ears and stands on the wet rooftop from image 3 at night, eyes closed and smiling, one hand touching an ear cup. Keep his face, hair and cream cardigan exactly as in image 1, and keep the sand-beige and gold design of the headphones exactly as in image 2. The string lights and city skyline glow behind him, reflected in the wet floor. Add a clean white sans-serif headline at the top that reads "HEAR THE CITY".
Why it works. Three sources, each given a role by number, with the parts that must not change listed. The headline was added in the same step and came out spelled correctly.
This mode starts from your own image: add it in the playground before you run.
Try this prompt
Image editing
Noon street to rainy neon night
Turn this sunny noon street into the same street on a rainy night: wet pavement reflecting out-of-focus red and teal neon lights, light rain falling, a dark sky. Relight the woman to match, with teal light on one side of her face and warm red light on the other, and raindrops on the shoulders of her blazer. Keep her face, hair, earrings, tan blazer, white blouse and pose exactly the same.
Why it works. One source, keeping its shape. The setting and the light change completely while the person, outfit and pose are named as fixed. The neon came out mostly red on her face, with the teal in the background.
This mode starts from your own image: add it in the playground before you run.
Try this prompt

Pricing

FLUX.2 Pro API pricing

One image starts at $0.03 at 1K and $0.039 at 2K.

The playground shows the exact price for your settings before you run. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.

See every price tier

Get started

How to use the FLUX.2 Pro API in three steps

Step 1
Write it like a photo brief
Subject, clothing, setting, light and lens in plain sentences. FLUX.2 reads long, specific prompts well, up to 5,000 characters.
Step 2
Quote text and pin colours
Put every word that must appear in quotes and say where it sits. Add hex codes such as #1B3A4B for the colours that matter.
Step 3
Number your references
In edits, refer to the sources as image 1, image 2 and image 3 in the order you upload them, and say what must stay exactly the same.

Open the playground Get an API key

Parameter reference and code samples are in the API and Readme tabs at the top of this page.

Why SpicyAPI

Why run FLUX.2 Pro on SpicyAPI

No filter from us
We add no content filter of our own. What the model renders is what you get back.
Tested, not guessed
Content labels come from real generations we ran and checked, with the test date on the page.
Pay per result, in dollars
No subscription and no credits. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.
One key for every model
One API key and balance cover image, video, audio and chat models, so you can explore more models from the same account.

FAQ

FLUX.2 Pro API FAQ

About FLUX.2 Pro

FLUX.2 [pro] is Black Forest Labs' production model for photoreal portraits, posters with readable text and colours pinned by hex code, with up to three references in an edit. Moderation applies, so it serves campaign, editorial and character work rather than the strongest content levels.