How is Base different from the distilled FLUX 3 Klein?

Both are the same 5.6B model. Black Forest Labs step-distilled one checkpoint to run in 16 steps for fast generation and publishes the undistilled Base, run at 50 steps with guidance 3.5, for fine-tuning, LoRA training and research. Here Base took about two to three times as long per image and toned down more of our stronger content tests than the distilled build.

How do I refer to several images in an edit?

Write image 1, image 2 and image 3 in the order you uploaded them, and say what each one brings: the woman from image 1, the dress from image 2. The first image also sets the shape when the aspect ratio is auto.

Will it render the strongest NSFW content?

Often not. The platform adds no filter, but this undistilled variant softened most prompts from the Strong level up in our tests and kept the Very strong scene implied; the Mild and Moderate levels rendered as asked. For the highest levels, use the distilled FLUX 3 Klein.

Can I load my own LoRA?

Not on this endpoint. FLUX 3 Klein Base LoRA runs the same model with up to three LoRA files per request.

Per image. 2k costs about four times 1k, and editing adds a charge for every source image. Aspect ratio and output format do not change the price. Failed requests are not charged.

FLUX 3 Klein Base API — Undistilled Image Model

black-forest-labs/flux-3-klein-base/text-to-image

FLUX 3 Klein Base text to image draws one picture from a written caption on Black Forest Labs' undistilled FLUX 3 Image Klein. Fifteen aspect ratios at 1k or 2k, JPEG, PNG or WebP.

From $0.036/image

Run model View API

Input

Text to image

Prompt567/5000

Aspect ratio

Resolutionaffects price

Output format

JPEGPNGWebP

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-3-klein-base/text-to-image",
  "input": {
    "aspect_ratio": "2:3",
    "resolution": "2k",
    "output_format": "jpeg",
    "seed": 111,
    "prompt": "Cinematic photograph, 50mm lens at eye level. A Scandinavian woman in her early thirties with wet ash-blonde hair combed back sits on the upper pine bench of a dim wood-fired sauna lounge, wearing a loose white linen shirt dress unbuttoned at the collar and slipping off one shoulder, knees drawn up, giving the camera a slow, smouldering glance. A narrow blade of cold blue daylight falls between the pine boards across her face and damp shoulder; the stove in the corner glows faint amber. Muted teal and honey colour grade, fine film grain, realistic skin texture."
  }
}

Introduction

FLUX 3 Klein Base Text to image draws one image from a written prompt on Black Forest Labs' FLUX 3 Image Klein Base, the undistilled 5.6B open-weights checkpoint released on 15 October 2026. Its makers run Base at 50 sampling steps against 16 for the distilled Klein, so each picture takes longer. Choose one of fifteen aspect ratios at 1k or 2k and JPEG, PNG or WebP. The platform adds no filter of its own, but this variant often softens the strongest prompts; for those, use the distilled FLUX 3 Klein.

Choose a mode by starting material

What you have	Mode
a written description	FLUX 3 Klein Base Text to image
one to three images to change or combine	FLUX 3 Klein Base Image editing

Key features & highlights

Full sampling, natural light

Without any shortcut in sampling, Base drew soft, believable skin and mixed light well in our tests: cold blue daylight through pine boards against an amber stove glow, wet hair, wool and linen textures. It suits stills you will look at closely rather than quick drafts, and the 2k tier gives about four megapixels to crop from.

Fifteen shapes

Every ratio from 3:1 to 1:3 is available, including 21:9, 4:5 and 2:1, and each came back at its own native size in our tests, from 1760 x 576 to 576 x 1760 at 1k. Banners, posters and phone screens come straight out of the model without cropping.

Repeatable seeds

A repeated seed with the same inputs returned identical pixels in our tests. Change one sentence at a time and every difference you see comes from that change, which makes prompt work on a slow model far less of a guess.

Request parameters

Request parameters for FLUX 3 Klein Base · Text to Image
Parameter	Required	Description
`prompt`	Yes	Required, up to 5,000 characters. Write a full caption: framing, subject, clothing, setting, light and lens. at least 1 characters · at most 5000 characters
`aspect_ratio`	No	Fifteen shapes from 3:1 to 1:3; 1:1 by default. At 1k, 16:9 came back as 1360 x 768 and 1:3 as 576 x 1760. `1:1` · `16:9` · `9:16` · `4:3` · `3:4` · `3:2` · `2:3` · `4:5` · `5:4` · `21:9` · `9:21` · `2:1` · `1:2` · `3:1` · `1:3` · Default: `1:1`
`resolution`	No	1k (default) is about one megapixel; 2k is about four megapixels, for example 2720 x 1536 for a wide picture. 2k costs about four times 1k and takes about a minute. `1k` · `2k` · Default: `1k`
`output_format`	No	JPEG (default), PNG or WebP. Does not change the price. `jpeg` · `png` · `webp` · Default: `jpeg`
`seed`	No	Optional, 0 to 2,147,483,647. In our tests the same seed and inputs returned identical pixels, so it isolates the effect of one change. `≥ 0` · `≤ 2147483647`

Pricing

Billed per image. 2k costs about four times 1k; the aspect ratio and output format do not change the price.

Pricing
Specification	Price
`1k`	$0.036 / image
`2k`	$0.144 / image

Use cases

Editorial portraits and interiors

Sauna lounges, cabins, workshops and lamp-lit interiors with layered light and fabric, when you want the model to take its time over skin and texture and you can wait about half a minute for each frame.

Before you fine-tune

Base is the checkpoint Black Forest Labs publishes for fine-tuning. See how it draws your subject and style from a prompt before you decide whether a LoRA or a full fine-tune is worth training.

Explore key use-case tutorials

Prompt guide & pro tips

Write captions, not keywords

Klein is trained on long scene captions, and Black Forest Labs normally rewrites short prompts before generation. This endpoint sends your words as written, so describe framing, subject, clothing, place, light, lens and grade in full sentences.

Name every garment

For stronger requests the model makes its own wardrobe choices. Put your words into the setting, the light and the lens instead, and let a reference image carry the outfit when it matters.

Draft at 1k, finish at 2k

1k takes about 14 to 45 seconds and costs a quarter of 2k. Settle the prompt and seed at 1k, then render the keeper at 2k for about four megapixels.

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-3-klein-base/text-to-image",
  "input": {
    "prompt": "A ceramic teapot on a linen cloth, soft window light, muted palette",
    "resolution": "1k",
    "aspect_ratio": "1:1",
    "output_format": "jpeg"
  }
}'

# 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-WEIGHTS IMAGE API

FLUX 3 Klein Base API at a glance

The undistilled FLUX 3 Image Klein: slower, full 50-step sampling, text to image and multi-image editing at up to four megapixels.

Open-weights model: 5.6B; FLUX 3 Image Klein by Black Forest Labs, released 15 October 2026.
Undistilled: 50 steps; Base keeps full sampling; the distilled Klein runs 16 steps for speed.
Aspect ratios: 15; From 3:1 to 1:3, including 21:9, 4:5 and 2:1.
Largest size tier: 4 MP; 2k is about four megapixels, such as 2720 x 1536; 1k is about one megapixel.
Edit sources: 3; Combine up to three images and refer to them as image 1, image 2 and image 3.
Softens strong prompts: Often; Prompts from the Strong level up mostly came back toned down; the distilled FLUX 3 Klein does not hold back like this.

Results

What you can make with FLUX 3 Klein Base

Text to image
A full caption, rendered as written
Klein is trained on long scene captions. Describe the light, the fabric and the pose, and Base draws each of them at up to four megapixels.
Text to image, 2:3 at 2k.
Two sources
A person from one photo, a dress from another
Name the sources as image 1 and image 2: the woman from a winter street photo, the velvet dress from a product shot, together in a sauna changing room.
Edit combining a portrait and a dress photo.
Edit at 2k
From an office to a cabin
One office portrait became a reclining pose in a cold-lit cabin at 2720 by 1536, with the same face and smile.
Edit of one office portrait, 16:9 at 2k.

More clips made with FLUX 3 Klein Base

Text to image, 4:5 at 1k.
Edit of one outdoor portrait, 4:5.
Text to image, 16:9 at 1k.

Content test

Is FLUX 3 Klein Base 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 3 Klein Base, by mode
Level	Text to image	Image editing
MildThe lightest prompts in the test	Not tested yet	Not tested yet
ModerateWhere filtered models start to refuse	Not tested yet	Not tested yet
StrongThe middle of the scale	SoftenedTested Oct 16, 2026	SoftenedTested Oct 16, 2026
Very strongWhere most models start to hold back	SoftenedTested Oct 16, 2026	SoftenedTested Oct 16, 2026
MaximumThe top of the scale	SoftenedTested Oct 16, 2026	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: The platform adds no content filter, but this undistilled variant often tones results down. In our tests it drew the Mild and Moderate levels as asked, yet softened most prompts from the Strong level up, toned down a male subject, kept the Very strong scene implied, and in editing returned the source unchanged. If you need the highest levels, use the distilled FLUX 3 Klein instead.
All modes: It is slower than the distilled FLUX 3 Klein: Base runs about 50 sampling steps where the distilled build runs 16. In our tests a 1k image took about 14 to 45 seconds of model time, a 2k image about a minute, and a 2k edit close to 90 seconds.
All modes: There is no negative prompt and no step or guidance control on this endpoint. One image per request, as JPEG, PNG or WebP.
Image editing: Editing takes one to three sources. With auto the result keeps the first image's shape at the chosen size tier; pick a ratio to change the shape. Each extra source adds to the price.
All modes: Every person you describe or upload must be an adult. You are responsible for the lawfulness of what you generate and publish, including the rights of anyone shown in a source image. Generated media is retained for 14 days and then deleted.

Use cases

What to build with the FLUX 3 Klein Base API

Considered stills
Portraits and interiors where you can wait half a minute for each picture and want full sampling rather than a fast preview.
Wardrobe and scene changes
Keep a person and change outfit, place and light in one edit, or dress them in a garment from a product photo.
Checking a fine-tune target
Base is the checkpoint Black Forest Labs publishes for fine-tuning; see how it handles your subject before you train on it.
Wide and tall formats
Banners at 3:1 or 21:9 and phone screens at 9:16 or 1:3 come straight from the model, no cropping needed.

Prompts

FLUX 3 Klein Base prompt examples you can copy

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

Text to image
Linen and a blade of daylight
Cinematic photograph, 50mm lens at eye level. A Scandinavian woman in her early thirties with wet ash-blonde hair combed back sits on the upper pine bench of a dim wood-fired sauna lounge, wearing a loose white linen shirt dress unbuttoned at the collar and slipping off one shoulder, knees drawn up, giving the camera a slow, smouldering glance. A narrow blade of cold blue daylight falls between the pine boards across her face and damp shoulder; the stove in the corner glows faint amber. Muted teal and honey colour grade, fine film grain, realistic skin texture.
Why it works. It reads like a caption: lens, pose, garment, two light sources and a grade, which is what Klein is trained on.
Try this prompt
Image editing
Dress from a product shot
Dress the woman from image 1 in the burgundy velvet slip dress from image 2. Keep her face, dark wavy hair and build. Move her into a sauna changing room with pale spruce walls: she leans back against the wall with damp hair, bare shoulders, one thin strap slipping off, and an alluring sideways glance. Cold blue daylight from above, faint amber stove glow, cinematic photograph, 50mm lens, fine grain.
Why it works. Each source gets a role by number, then the instruction names what stays and describes the new place and light.
This mode starts from your own image: add it in the playground before you run.
Try this prompt
Text to image
Two people, two poses
Intimate cinematic photograph, 35mm lens. In the loft of a log cabin at blue dusk, a Latina woman around thirty with long dark hair, in an oversized oatmeal knit sweater dress, sits on a sheepskin rug beside a black cast-iron wood stove, leaning back against the chest of a fair-haired man in his late thirties in a navy wool sweater who sits behind her with his arms around her waist. Both faces are turned toward the stove, her eyes half closed, a soft smile. Cold blue twilight fills the room while the stove throws warm orange light across their faces. Soft contrast, gentle grain, realistic skin texture.
Why it works. Giving each person a position and a direction to look keeps both faces clean in a close embrace.
Try this prompt

Pricing

FLUX 3 Klein Base API pricing

One image starts at $0.036 at 1K and $0.144 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 3 Klein Base API in three steps

Step 1
Write a full scene caption
Klein is trained on long captions. Describe framing, subject, clothing, setting, light and lens in sentences; a one-line prompt gives a flatter picture.
Step 2
Pick shape and size
Choose one of fifteen aspect ratios and 1k or 2k. Work at 1k while you settle the prompt; 2k costs about four times as much and takes about a minute.
Step 3
Edit with up to three images
Upload sources in the order you name them and write "image 1", "image 2" in the instruction, for example the woman from image 1 in the dress from image 2.

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 3 Klein Base 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 3 Klein Base API FAQ

About FLUX 3 Klein Base

FLUX 3 Klein Base is the undistilled FLUX 3 Image Klein: full 50-step sampling for text to image and editing with up to three images, and fifteen shapes at up to four megapixels. Choose it when you can trade speed for a considered render.