Is there a content filter on NSFW images?

No. In our tests nothing was refused, from Mild to Maximum. The limit is the model: it often adds garments on its own, and it does not deliver the Very strong level.

How is it different from FLUX.2 klein 4B?

Same family and the same fast workflow, one size up. In our tests the 4B was the most cautious of the open FLUX.2 models, often adding garments on its own, and its hands often went wrong with two people in frame. The 9B is the larger of the two, so it is the one to start with when you want fewer cover-ups and steadier results.

How is it different from the larger FLUX.2 models?

It is much faster and cheaper per image, and it generates and edits in the same model. In exchange, text and fine detail are less reliable than on FLUX.2 [pro] or [max]. Those two are filtered by Black Forest Labs; this one is not.

Can it write text into images?

Short, quoted text usually works: our lookbook render spelled its header and all four captions correctly, and a neon sign came out clean on the third seed we tried. It is not reliable every time, so plan to try a few seeds.

Does it follow hex colour codes?

Roughly. A requested sand-coloured background came out in the right direction but not as the exact value, so do not use it where a precise brand colour is required.

Can I edit photos other editors refuse?

Editing does not refuse. In our tests it kept the person and changed clothes and scenes, and an unfiltered source was accepted and kept as it was. Only the Very strong level came back toned down.

How many reference images can I use?

One to three in an edit. Mention them in your prompt in the order you uploaded them.

Text to image is billed per image, and 2k costs more than 1k; the aspect ratio does not change the price. An edit is one flat price per image whatever the number of sources or the size tier. Failed requests are not charged.

FLUX.2 Klein 9B API: Fast Uncensored NSFW Images

black-forest-labs/flux-2-klein-9b/text-to-image

FLUX.2 [klein] 9B text to image turns a written prompt into a 1k or 2k picture in about half a minute, in nine aspect ratios. There is no content filter, but the small model often dresses its subjects.

From $0.011/image

Run model View API

Input

Text to image

Prompt854/5000

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-klein-9b/text-to-image",
  "input": {
    "aspect_ratio": "3:2",
    "resolution": "1k",
    "seed": 260121,
    "prompt": "A four-panel fashion lookbook contact sheet on a warm sand backdrop (#E8D5C4). The same woman in her thirties, with a sleek dark bob and olive skin, appears in all four panels with an identical face. Panel 1: champagne satin slip dress with thin straps and a lined bodice, one strap slipping off her shoulder. Panel 2: oversized white linen shirt worn open over a black bandeau top and high-waisted shorts. Panel 3: backless black halter jumpsuit, photographed from behind as she looks back over her shoulder. Panel 4: red one-piece swimsuit with a scoop neckline and high-cut legs, a sheer red sarong tied at the hip, one hand resting on her hip. Soft studio strobe light, editorial fashion photography. A clean bold sans-serif header across the top reads \"RESORT EDIT 26\", and each panel has a small caption: \"LOOK 01\", \"LOOK 02\", \"LOOK 03\", \"LOOK 04\"."
  }
}

Introduction

FLUX.2 [klein] 9B Text to image is Black Forest Labs' small, fast image model, released on 15 January 2026, generating pictures from a written prompt. It renders at 1k or 2k in nine aspect ratios, usually in 15 to 35 seconds, and it has no content filter.

Choose a mode by starting material

What you have	Mode
a written brief	FLUX.2 Klein 9B Text to image
images to change	FLUX.2 Klein 9B Edit

Key features & highlights

Fast enough to iterate

Most of our renders finished in 15 to 35 seconds, so trying five wordings or five seeds costs little time. It is the quick end of the FLUX.2 line.

Text and look sheets

Quote a header or a sign and it is usually rendered as written. One prompt can also lay out several looks of the same person in a grid. Neither is guaranteed on every seed.

No filter, honest limits

Nothing is refused. The limit is the model: it often adds garments on its own, and it does not deliver the Very strong level.

Request parameters

Request parameters for FLUX.2 Klein 9B · Text to Image
Parameter	Required	Description
`prompt`	Yes	Required, up to 5,000 characters. Describe subject, clothing, setting, light and style in plain sentences; put words that must appear in the image in quotes. at least 1 characters · at most 5000 characters
`aspect_ratio`	No	One of nine shapes from 21:9 to 9:21; 1:1 by default. 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, about one megapixel) or 2k (2048 pixels on the long side). 2k costs more than 1k. `1k` · `2k` · Default: `1k`
`seed`	No	Optional whole number from 0 to 2,147,483,647. Reuse it to compare prompt changes side by side; leave it empty for a random one. `≥ 0` · `≤ 2147483647`

Pricing

Billed per image by size tier: 2k costs more than 1k, and the aspect ratio does not change the price. Failed or refused requests are not charged.

Pricing
Specification	Price
`1k`	$0.011 / image
`2k`	$0.0122 / image

Use cases

Batches of variations

Many takes on one idea, to pick from or to feed into an edit.

Character look sheets

Several outfits of one character in a single frame.

Drafts for larger models

Settle on composition here before spending on a bigger model.

Explore key use-case tutorials

Prompt guide & pro tips

Say what the people wear

Garment, cut and how it sits. Without it the model fills in clothing by itself.

Quote text and keep it short

A few words per line render best. If a word is wrong, change the seed.

Do not rely on exact hex colours

A hex value gets you the right direction, not the exact shade.

Start at 1k

It is the cheaper tier; move to 2k for the frame you keep.

Try three seeds

Small models vary. Hands and fine text improve with a second or third try.

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-klein-9b/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.

OPEN-WEIGHT FAST IMAGE MODEL

FLUX.2 Klein 9B API at a glance

FLUX.2 [klein] 9B from Black Forest Labs: a small, fast image model that generates and edits in one, with no content filter.

Typical render: 15–35 s; Most images in our tests finished in 15 to 35 seconds, much faster than the larger FLUX.2 models.
Reference images: 3; Edit with up to three sources, in the order your prompt refers to them.
Largest size tier: 2k; 2048 pixels on the long side; 1k is about one megapixel.
Aspect ratios: 9; From 21:9 to 9:21, in both text to image and editing.
Content filter: None; No review of prompts or results: nothing is refused.

Results

What you can make with FLUX.2 Klein 9B

Look sheets
Four outfits, one woman, one render
A single prompt produced a four-panel lookbook with a bold header, four captions spelled as written, and the same woman in four different outfits. A stray cream strip runs under the panels, a typical small-model slip.
Text to image, 2k, 3:2.
Text in the scene
A neon sign that reads correctly
A couple in a late-night laundromat under flickering tubes, with a pink sign in the window reading OPEN LATE. It took a few seeds to get the sign clean; the pose differs from the prompt, with her legs uncrossed.
Text to image, 2k, 4:3.
Edit
New shirt, new room, same man
One source photo of a man against a white wall became him in an open black silk shirt in a dim jazz bar. His face, beard and hands-in-pockets stance carried over.
Edit, one source, auto shape.
Edit at 2k
A cafe turned into a rooftop at night
Two women laughing at a daytime cafe table, moved to a rooftop under string lights with a skyline behind them, new dresses and cocktails instead of coffee, rendered at 2048 by 1152.
Edit, one source, 16:9 at 2k.

Content test

Is FLUX.2 Klein 9B 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 Klein 9B, by mode
Level	Image editing	Text to image
MildThe lightest prompts in the test	RendersTested Oct 8, 2026	RendersTested Oct 8, 2026
ModerateWhere filtered models start to refuse	RendersTested Oct 8, 2026	RendersTested Oct 8, 2026
StrongThe middle of the scale	RendersTested Oct 8, 2026	Not tested yet
Very strongWhere most models start to hold back	Not tested yet	Not tested yet
MaximumThe top of the scale	RendersTested Oct 8, 2026	RendersTested Oct 8, 2026

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: There is no content filter on this model: in our tests prompts at every level from Mild to Maximum were accepted and nothing was refused. That is not the same as getting every level delivered. In text to image this model often adds garments on its own to Strong-level prompts, while edits accepted an unfiltered source image and kept it as it was; the Very strong level came back toned down in both modes. The larger FLUX.2 [dev] draws less cautiously; the smaller klein 4B is more cautious still.
All modes: It is a small model. Text in the image and colours written as hex codes are not reliable: a quoted headline came out clean in some runs and garbled in others, and a requested hex background came out close in tone but not exact. Hands and fine detail can slip. Try a few seeds. Output is JPEG, one image per request.
Image editing: Editing takes one to three source images. With auto the result keeps the shape of the first image at about one megapixel; choose an aspect ratio to get the 1k or 2k size tier instead. In our tests the face and pose of a person were kept through outfit and background changes, with no mask or inpainting.

Use cases

What to build with the FLUX.2 Klein 9B API

Look sheets and batches
A grid of outfits or poses for one character in a single render, or dozens of variations of one idea in a short session.
Quick scene and outfit swaps
Keep a person and change the clothes, the room or the time of day in an edit that returns in well under a minute.
Drafts before a larger model
Settle on composition and mood here, then move to a bigger FLUX.2 model for the final frame.

Prompts

FLUX.2 Klein 9B prompt examples you can copy

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

Text to image
A lookbook with a header and captions
A four-panel fashion lookbook contact sheet on a warm sand backdrop (#E8D5C4). The same woman in her thirties, with a sleek dark bob and olive skin, appears in all four panels with an identical face. Panel 1: champagne satin slip dress with thin straps and a lined bodice, one strap slipping off her shoulder. Panel 2: oversized white linen shirt worn open over a black bandeau top and high-waisted shorts. Panel 3: backless black halter jumpsuit, photographed from behind as she looks back over her shoulder. Panel 4: red one-piece swimsuit with a scoop neckline and high-cut legs, a sheer red sarong tied at the hip, one hand resting on her hip. Soft studio strobe light, editorial fashion photography. A clean bold sans-serif header across the top reads "RESORT EDIT 26", and each panel has a small caption: "LOOK 01", "LOOK 02", "LOOK 03", "LOOK 04".
Why it works. Four looks of one woman in a single frame, with a quoted header and quoted captions. The text and the identical face held; the background colour came out close, not exact.
Try this prompt
Text to image
A neon sign in a late-night scene
Late night in a 24-hour laundromat, a couple in their thirties. She sits on top of a running dryer in an oversized flannel shirt slipping off one shoulder over denim cutoff shorts, bare legs crossed; he stands close in a white tank top and grey sweatpants, one hand on the dryer beside her, their foreheads almost touching. Flickering fluorescent tubes overhead, a pink neon sign in the rain-streaked window reads "OPEN LATE", rows of washing machines behind them. 35mm film photograph, fine grain, shallow depth of field.
Why it works. Pose, clothing, light and film look are spelled out, and the sign text is quoted. The sign came out clean on a later try, so be ready to change the seed.
Try this prompt
Text to image
Two men on a fire escape
Two men in their thirties on a city fire escape on a hot summer evening. One is shirtless in low-slung jeans, leaning back against the brick wall with his eyes half closed; the other, in an unbuttoned short-sleeve shirt with a palm-leaf print, sits on the step below him and passes him a cold bottle of beer, their knees touching, easy smiles. Orange sunset haze between the buildings, potted basil on the iron railing, warm film color, candid documentary photography.
Why it works. Clothing, sunset light and a small shared action give the model enough to build a relaxed documentary scene; the seating came out a little different from what was written.
Try this prompt
Image editing
New outfit and room, same person
Keep the same man from image 1: same face, beard, grey hair and pose, hands in his pockets. Change his grey t-shirt to an unbuttoned black silk shirt worn open over his chest, with tailored charcoal trousers. Replace the white wall with a dim jazz bar: amber table lamps, a brass bar counter and rows of glowing bottles behind him, a soft haze in the warm light. Keep the photo realistic.
Why it works. The prompt lists what stays (face, beard, hair, pose) before the change. The man is recognisable, and the shirt and room are clearly new.
This mode starts from your own image: add it in the playground before you run.
Try this prompt
Image editing
Day cafe to rooftop night
Keep the same two women, their faces, hair and seated pose, laughing together. Turn the sunny cafe into a rooftop party at night: warm string lights overhead, a city skyline glowing behind them, cocktail glasses on the table instead of coffee. Dress the woman on the left in a black satin slip dress with thin straps and the woman on the right in a white silk camisole with a delicate gold necklace. Candlelit warm glow on their skin, realistic photo.
Why it works. Both women are kept in their seats and expressions while the place, the light and their outfits change, and the result is large enough to print small.
This mode starts from your own image: add it in the playground before you run.
Try this prompt

Pricing

FLUX.2 Klein 9B API pricing

One image starts at $0.011 at 1K and $0.012 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 Klein 9B API in three steps

Step 1
Use it for volume and iteration
Renders are fast and cheap enough to try several wordings or seeds, then keep the best. Small models vary more from seed to seed than large ones.
Step 2
Quote short text and retry
Short headers and signs usually render, but not every time. If a word comes out wrong, change the seed instead of rewriting the prompt.
Step 3
Describe the outfit you want
The model tends to add clothing on its own. Name the garments, their cut and how they sit on the body to steer it, and say what must stay the same when you edit.

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 Klein 9B 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 Klein 9B API FAQ

About FLUX.2 Klein 9B

FLUX.2 [klein] 9B is the fast end of the FLUX.2 line: it renders in about half a minute, edits with up to three references, and has no content filter. It is a small model, so expect to try a few seeds, and expect it to dress its subjects more often than the larger ones.