Skip to main content

FLUX.2 [dev] image editing from Black Forest Labs' open-weights model. It takes one to three source images and a prompt, keeps faces recognisable, and can combine people from separate photos into one scene.

From $0.023/image

Workbench

Input

Image editing
637/5000

Reordering updates prompt references. Removing media also removes its references; review your prompt.

Up to 5,000 characters. Say what to change and what must stay the same. When you combine photos, name each person by image number and ask to keep their faces. A hex code such as #D98E04 steers the colour of new light. at least 1 characters · at most 5000 characters
2/3

Drop files or media links, paste, or browse · Up to 3 files · Max 10 MB · JPG · PNG · WEBP · GIF

  • image 1
  • image 2
2 / 3 files
One to three source images, up to 10 MB each, in the order your prompt refers to them ("image 1", "image 2"). Clear, front-facing photos keep faces best. at least 1 items · at most 3 items
auto (default) follows the shape of the first source image at about one megapixel. Choose one of nine ratios from 21:9 to 9:21 to recrop. It does not change the price.
1k (default) or 2k, used only when you choose a specific aspect ratio. With auto the result stays at about one megapixel. The price is the same.
Seed
A whole number from 0 to 2,147,483,647. Keep it fixed while you adjust the prompt so results are easy to compare; it does not guarantee an identical image. It does not change the price. no lower than 0 · no higher than 2147483647
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
two-portraits-cellar-danceExample result

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 OPEN-WEIGHTS IMAGE MODEL

FLUX.2 Dev API at a glance

FLUX.2 [dev]: the open-weights FLUX.2 model, photoreal portraits and readable text, with no content filter.

Parameters
32B
Black Forest Labs' FLUX.2 [dev], released with open weights on 25 November 2025.
Reference images
3
Edit mode takes one to three source images and can combine people from separate photos.
Long side at 2k
2048 px
1k is about one megapixel; 2k reaches 2048 pixels on the long side.
Aspect ratios
9
From 21:9 to 9:21, with the same price at every ratio.
Colour codes
Hex
Write a value such as #E9A23B in the prompt and the model steers the colour of light and objects towards it.
Content filter
None
Uncensored: no review of the prompt or the result, and no refusals in our tests.

Results

What you can make with FLUX.2 Dev

  • Two women in their late thirties laughing together at a wooden counter in a vintage record store, a sign reading CLOSED on the glass door behind them

    Text and colour

    A sign that reads CLOSED

    Two women in a record store after hours, a quoted sign on the glass door and a warm #E9A23B lamp glow. The sign is spelled correctly and the light follows the colour code.

    Text to image, 2k, 3:2.

  • A woman with damp auburn hair in a motel room, a pink VACANCY neon sign glowing in the window behind her

    Portraits

    Neon from the window

    A woman in a motel room at night, a "VACANCY" sign outside, magenta and blue light mixed across the room. It comes back with visible film grain.

    Text to image, 2k, 2:3.

  • A bearded man with an open white shirt and a loose navy tie leaning on a dark wood hotel bar with a glass of whiskey, smiling at the camera

    Editing

    Same man, a different night

    A buttoned-up office portrait becomes a late hotel bar with the shirt open and the tie loosened. The face and beard stay, the room and mood change.

    Edit, one source, auto shape.

  • A woman in a midnight-blue silk dress and a man in an open white shirt dancing close together in a stone wine cellar lit by candles

    Two references

    Two portraits, one dance

    A woman from one photo and a man from another, slow-dancing in a candlelit cellar. Both faces stay close to their sources.

    Edit, two sources, 2k, 3:2.

Content test

Is FLUX.2 Dev 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 Dev, by mode
LevelText to imageImage editing
MildThe lightest prompts in the test
RendersTested Oct 8, 2026
RendersTested Oct 8, 2026
ModerateWhere filtered models start to refuse
RendersTested Oct 8, 2026
Not tested yet
StrongThe middle of the scale
RendersTested Oct 8, 2026
RendersTested Oct 8, 2026
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
No filter is applied and requests are not refused. The limits come from the model itself: in our tests text to image delivered the Mild, Moderate, Strong and Maximum levels in every run, but the Very strong level came back toned down, implied rather than shown, and male subjects came out rough.
Text to image
One image per request, up to 2048 pixels on the long side. There is no negative prompt, no step count and no guidance control, and the result is a JPEG. Long passages of text can still slip, so keep it to a few words.
Image editing
One to three source images of up to 10 MB each. Faces carry over well from clear, front-facing photos but can drift on small or tilted ones, and a second person's likeness is a little less stable than the first.

Use cases

What to build with the FLUX.2 Dev API

  • Portraits and lifestyle scenes

    Warm, grainy film-style stills of one or two people in a believable place: record stores, motel rooms, sailboat decks. Faces, fabric and light read as photographs.

  • Posters, signs and labels

    Short pieces of text such as shop signs, neon lettering and labels render legibly, so mock-ups need less retouching.

  • Outfit, scene and mood changes

    Send a portrait and ask for a different outfit, setting and lighting while the person stays recognisable.

  • Bringing two people into one frame

    Give it two separate portraits and ask for a shared scene. It keeps both faces close to the photos.

  • Unfiltered work

    There is no filter on the prompt or on the result, which suits platforms whose users expect that freedom. In our tests it delivered every intensity level except Very strong, which came back toned down.

Prompts

FLUX.2 Dev prompt examples you can copy

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

  • Text to image

    Two women, a quoted sign and warm lamps

    Photograph of two women in their late thirties in a small vintage record store after closing time. One, with a cropped black pixie cut, a few grey strands and gold hoop earrings, sits on the wooden counter wearing a loose oatmeal cardigan slipping off one shoulder over a thin black camisole and high-waisted jeans; the other, a Black woman with a short natural afro and laugh lines, stands close beside her in a fitted burgundy turtleneck. They hold a vinyl record sleeve between them, foreheads almost touching, laughing softly. Crates of records, warm #E9A23B tungsten lamps, a small sign hanging on the glass door reads "CLOSED". 35mm film photo, soft grain, shallow depth of field.

    Why it works. Clothing, hair and the shared record are spelled out, the sign text is in quotes and the lamp colour is a hex code. The sign reads CLOSED and the warm glow follows the colour.

    Try this prompt
  • Text to image

    Neon light in a motel room

    Photograph of a woman in her mid-thirties with damp, tousled auburn hair sitting on the edge of a motel bed at night, wearing a champagne silk slip dress with one thin strap slipped off her shoulder, legs crossed, looking toward the camera with a calm, knowing expression. Through the window behind her, a pink neon sign reads "VACANCY". The sign washes the room in #FF4F8B magenta mixed with cool #3AA7FF blue. 35mm film photo, visible grain, slightly soft focus at the edges.

    Why it works. Two hex colours for the light, a quoted sign and a film look. The room takes on the magenta and blue mix and the sign is spelled correctly.

    Try this prompt
  • Text to image

    Two men on a sailboat deck

    Photograph of two men in their late thirties on the teak deck of a sailboat at golden hour. One, shirtless and sun-tanned with a short dark beard, sits on the deck coiling a rope; the other, in an unbuttoned white linen shirt and navy swim shorts, leans against the mast laughing and holds out a plain brown glass bottle to him. Low warm sun flaring behind the sail, sea in #2E6F95, light sea spray, candid documentary style, 35mm.

    Why it works. A candid documentary scene: low sun, sea colour and what each man is doing. Both read as relaxed and natural in the golden light.

    Try this prompt
  • Image editing

    The same man in a hotel bar

    Same man, same face and beard, alone in the frame. It is now late at night after work: his suit jacket is gone, his navy tie hangs loose and untied around his neck, and his white shirt is fully unbuttoned and hanging open over his bare chest, sleeves rolled to the elbows. Replace the bright office with a dim hotel bar of dark wood and warm #D98E04 amber lamps. He leans one elbow on the bar holding a glass of whiskey and gives a slow, confident smile toward the camera. Realistic photograph, shallow depth of field. No other people.

    Why it works. One portrait, and the prompt lists what changes in the clothes and the room. The face and beard stay while the whole mood shifts.

    This mode starts from your own image: add it in the playground before you run.

    Try this prompt
  • Image editing

    Two separate portraits, one scene

    Combine the two people into one scene: the woman from image 1 and the man from image 2 slow-dancing close together in an old stone wine cellar lit by candles. Keep both faces exactly as in the photos, with the same age, features and fine lines; her face is turned toward the camera, resting against his shoulder. She now wears a midnight-blue silk dress with thin straps and an open back, her hair still in a low bun with small gold hoop earrings; he wears a white dress shirt open at the collar with the sleeves rolled up, and dark trousers. His hand rests on her bare lower back. Warm #F4A259 candlelight, deep shadows, 35mm film look.

    Why it works. Each person is named by image number and the faces are marked as fixed. Both stay recognisable in the new candlelit setting.

    This mode starts from your own image: add it in the playground before you run.

    Try this prompt

Pricing

FLUX.2 Dev API pricing

One image starts at $0.012 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 Dev API in three steps

  1. Step 1

    Write it like a photo brief

    Name the subject and what they wear, the place, the light and the film or lens look in full sentences. FLUX.2 follows long, specific prompts well.

  2. Step 2

    Quote any text you want in the picture

    Put the exact words in quotation marks and say where they sit, such as a sign on a glass door. Our "CLOSED" and "VACANCY" signs came back spelled correctly.

  3. Step 3

    Colour the light with hex codes

    Add a code like #FF4F8B to the lamp or the neon you describe. It is a strong hint for the colour of that element, not a guarantee of an exact shade.

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

Why SpicyAPI

Why run FLUX.2 Dev 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 Dev API FAQ

About FLUX.2 Dev

FLUX.2 [dev] is the open-weights FLUX.2 model: photographic portraits, readable text, hex colour control and up to three reference images, in nine shapes up to 2048 pixels. There is no filter and no refusal. In our tests it delivered every level as asked except Very strong, which it toned down.

All FLUX models21

Image models20

Related models and next tasks