Head to head
Esc

Models · slot A

to navigate to pick

FLUX.1 [dev] Black Forest Labs FLUX.1 Krea [dev] Black Forest Labs

Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.

FLUX.1 [dev]

24.6 arena score

#16 of 62 in Text-to-Image

Skill signature · Text-to-Image

FLUX.1 Krea [dev]

18.8 arena score

#47 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 [dev]

0%

win rate

Ties

0%

FLUX.1 Krea [dev]

0%

win rate

Shared challenges 13

Challenge by challenge

The strongest take from each model on every shared challenge, with the AI judge's read.

Geometric Composition

Text-to-Image

“A glass cube on a wooden table. Inside the cube is a small blue sphere. On top of the cube sits a red book. A green plant is behind the cube, partially visible through the glass. Soft window light from the left.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to lighting instructions with a clear left-side window light source.
  • + Exceptional glass rendering with realistic thickness, refraction, and bevels.
  • + High-quality textures, particularly on the red leather-bound book.
  • The sphere appears to be floating mid-air rather than resting on the bottom surface of the cube.
  • The perspective of the plant leaves is slightly disjointed in the glass refraction.

FLUX.1 Krea [dev]

  • + Logical placement of the sphere resting on the internal floor of the cube.
  • + More ornate book detail with gold spine accents.
  • + Accurate spatial depth between the cube and the plant.
  • The glass box has odd vertical pillars inside that make it look more like a display case frame than a simple solid glass cube.
  • Lighting is a bit flatter and less dynamic than the source image.

Verdict: FLUX.1 [dev] produced a more visually striking image with superior lighting and materials, although the sphere is physically impossible (floating). FLUX.1 Krea [dev] handled the physical interaction of the objects more logically, but the geometry of the cube is somewhat cluttered by internal frames that aren't present in the prompt. FLUX.1 [dev] is the winner for its superior realism and high-fidelity rendering.

Candid Street Photography

Text-to-Image

“A candid street photo of an elderly Japanese man repairing a red bicycle in light rain, reflections on wet pavement, shallow depth of field, 50mm lens, natural skin texture, imperfect framing, motion blur from passing cars, cinematic but realistic, no stylization.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent high-frequency detail in the skin, hair, and clothing texture.
  • + Captures the 'light rain' atmosphere with visible droplets and realistic wet pavement reflections.
  • + Accurate 50mm lens feel with a natural shallow depth of field.
  • The man is holding the handlebars rather than performing a specific 'repair' action.
  • The framing feels a bit too clean and centered for the 'imperfect framing' request.

FLUX.1 Krea [dev]

  • + Successfully incorporates motion blur on passing cars as requested.
  • + The 'imperfect' composition feels more like a candid street snap.
  • + The bicycle design is realistic and appropriate for the context.
  • The man's right hand is fused with the bicycle seat in a significant anatomical error.
  • Skin and beard textures look slightly smoother and more 'AI-processed' compared to Model A.
  • The background cars have warped geometry and lack clarity.

Verdict: FLUX.1 [dev] produces a much higher quality image with superior skin textures and realistic lighting, though it misses the specific request for motion blur. While FLUX.1 Krea [dev] attempts more of the prompt details like motion blur and imperfect framing, it suffers from a major anatomical failure where the subject's hand merges with the bicycle.

Fantasy Warrior

Text-to-Image

“Close portrait of a battle-worn paladin in ornate engraved plate armor, hair braided with small beads, faint scars and dirt on the skin, warm torchlight reflecting off metal, shallow depth of field, bokeh sparks, lifelike eyes, highly detailed texture on leather straps and cloth underlayer.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Extremely high-quality facial details and lifelike iris textures.
  • + Dynamic warm lighting that creates a realistic glow on the skin and armor surfaces.
  • + Strong shallow depth of field effect with vibrant bokeh.
  • The character does not appear 'battle-worn', looking very clean with minimal dirt or scars.
  • Failed to include the beads in the braids requested in the prompt.
  • The armor lacks the specific 'ornate engraved' detail mentioned.

FLUX.1 Krea [dev]

  • + Excellent adherence to all prompt details including beads in hair, engraved armor, leather straps, and cloth underlayers.
  • + The character looks convincingly 'battle-worn' with visible grime and blood-like spatters.
  • + High technical detail in the complex engravings on the plate armor.
  • The lighting is a bit flatter and darker compared to the dynamic glow of the other image.
  • Skin texture is slightly less realistic than the other model's output.

Verdict: While FLUX.1 [dev] produces a more striking and lifelike portrait with superior skin and eye rendering, FLUX.1 Krea [dev] is the clear winner for prompt adherence. FLUX.1 Krea [dev] correctly included every requested detail—specifically the beads in the hair, the ornate engravings, and the battle-worn texture—which were mostly absent in FLUX.1 [dev].

Modern Clean Menu

Text-to-Image

“Modern minimalist restaurant menu design, white background with colorful food photos in grid, sections for appetizers/pizza/mains, bold sans-serif fonts, vibrant accents, clean professional layout for casual dining.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent legibility for header text and prices
  • + Very clean minimalist aesthetic with pleasant white space
  • + Coherent menu categories that align well with the prompt
  • Failed the grid requirement for photos
  • Slightly less 'vibrant' than requested
  • Text descriptions are largely gibberish upon close inspection

FLUX.1 Krea [dev]

  • + Successfully implemented the requested grid layout for photos
  • + Stronger visual interest with vibrant colors and bold headers
  • + Professional shadowing and lighting that makes it feel like a real product mockup
  • Text rendering is very messy and distorted across the whole image
  • Repeats the same category header ('Appetizers') unnecessarily
  • Layout feels a bit cramped compared to the other version

Verdict: FLUX.1 [dev] produced a much cleaner and more readable design that captures the 'minimalist' instruction perfectly, though it missed the specific request for a grid of photos. FLUX.1 Krea [dev] followed the structural layout prompt better with the grid, but the text rendering is significantly lower quality. FLUX.1 [dev] is the likely winner as its output is higher quality and more usable as a design template.

Magic Burger Explosion: Fiery Photorealism Challenge

Text-to-Image

“Ad for 'Magic Burger'. Dynamic, exploded burger with all components (bun, patty, cheese, lettuce, tomato, sauce) suspended in mid-air. Emphasize photorealistic detail and a sense of motion. Dark, fiery background with glowing embers. Integrate text: 'MAGIC BURGER' as a prominent title, 'LIMITED TIME ONLY' as a secondary message, and '€6.99' in a starburst, all rendered with a fiery, glowing effect.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent exploded view with clear separation of ingredients
  • + Clean and centered composition
  • + Accurate rendering of the secondary text and price
  • Missed the primary 'MAGIC BURGER' text entirely
  • The fire looks more like a small campfire at the bottom than a fiery background
  • Lacks the requested starburst for the price

FLUX.1 Krea [dev]

  • + Successfully included all requested text as prominent titles
  • + Better 'fiery background' with embers and glowing coals
  • + Included a starburst element for the secondary message
  • The burger is less 'exploded' into individual components compared to Model A
  • Text is not 'fiery and glowing' as requested, appearing more like standard graphic overlays
  • Price is not inside the starburst as specified

Verdict: FLUX.1 [dev] produced a superior exploded burger visualization with clear component separation, but failed significantly on the text requirements by omitting the main title. FLUX.1 Krea [dev] followed the text instructions more comprehensively and captured the fiery atmosphere much better, despite the burger being less dynamically disassembled. FLUX.1 Krea [dev] is the winner for following the complex prompt instructions more precisely.

Chalkboard Menu

Text-to-Image

“Handwritten-style chalkboard menu in a cozy café, all text rendered in the exact same realistic chalk handwriting style with natural variations in letter size, slight slant, and chalk texture — no printed or digital fonts anywhere on the board. Title at the top in elegant cursive chalk handwriting: ‘TODAY’S SPECIALS – APRIL 30, 2026’. Below it, three menu items also in the same handwritten chalk style: ‘Truffle Mushroom Risotto – $24’, ‘Grilled Octopus with Lemon & Herbs – $28’, ‘Brown Butter Chocolate Chip Cookies – $9’. At the very bottom, smaller text in the identical handwritten chalk style (slightly smaller but still clearly legible with the same handwriting characteristics): ‘All items made fresh daily • Ask about our gluten-free options’. Warm ambient café lighting, visible chalk dust and smudges, realistic handwriting imperfections, no clean printed text anywhere.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent text accuracy and legibility for all requested menu items.
  • + Consistent and realistic chalk texture across the entire board.
  • + Followed the specific item details and prices perfectly.
  • The 'cursive' for the title is more of a print-style sans-serif than elegant cursive.
  • Some minor spelling artifacts in the small bottom text like 'drish'.

FLUX.1 Krea [dev]

  • + Good chalk smudging and atmospheric lighting in the café setting.
  • + Generally follows the requested date and title format.
  • Repeated and hallucinated menu items that were not in the prompt, such as 'Gruffle Mushroom'.
  • Incorrect pricing and garbled words in the middle section of the list.
  • Lower text clarity compared to Model A.

Verdict: FLUX.1 [dev] is the clear winner as it followed the prompt instructions precisely, rendering all three specific menu items and their prices correctly. FLUX.1 Krea [dev] suffered from significant text hallucinations and repetition, resulting in a cluttered and nonsensical menu despite having a nice chalk aesthetic.

The Reversed Rodeo

Text-to-Image

“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent composition with a cinematic use of lighting.
  • + Clean rendering of the astronaut suit and horse anatomy.
  • + Matches the surreal request while maintaining high visual clarity.
  • The horse has an extra leg clearly visible at the bottom.
  • Ignored the specific 'horse on top' instruction, opting for the standard horse-under-rider pose.

FLUX.1 Krea [dev]

  • + Creative interpretation of the horse integrated with spacecraft technology.
  • + Sharp details on the planet surface and technical equipment.
  • + Good use of high-contrast lighting to enhance the cinematic feel.
  • Completely failed the negative constraint to have the 'horse on top'.
  • Anatomical issues where the horse's front legs merge awkwardly into the mechanical structure.

Verdict: Both models failed the specific prompt instruction to place the horse on top of the astronaut, providing a standard 'astronaut riding a horse' scene instead. FLUX.1 [dev] is the preferred image due to its superior lighting and more elegant aesthetic, despite an anatomical error with the legs, whereas FLUX.1 Krea [dev] produced a more cluttered and messy mechanical integration.

The Capybara Taxi Driver

Text-to-Image

“Photorealistic scene inside a yellow New York taxi at night. A capybara is driving, wearing a yellow taxi driver cap and a dark jacket. It has a calm, professional expression and both front paws on the steering wheel. In the back seat sits a human businesswoman in a coat, looking at her phone with a completely normal, bored expression (as if this is just another normal ride). Through the windows you can see the streets of Manhattan at night with blurred lights. Realistic taxi interior, photorealistic, detailed fur and fabric, 35mm lens, night lighting with reflections, shallow depth of field.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealistic texture on the capybara's fur
  • + High-quality rendering of the phone and female character
  • + Accurate moody lighting from the city street lights
  • The passenger is sitting in the front seat instead of the back seat as requested
  • The perspective makes the capybara appear strangely large relative to the car

FLUX.1 Krea [dev]

  • + Correctly places the woman in the back seat per the prompt
  • + Captures the bored expression of the passenger very well
  • + Realistic lateral view of the taxi cabin
  • The capybara's hands look more like primate hands than capybara paws
  • The capybara's head is clipping through the yellow hat
  • Random second face visible in the front passenger seat area

Verdict: Both models captured the surreal nature of the prompt well, but FLUX.1 Krea [dev] followed the layout instructions more accurately by placing the woman in the back seat. However, FLUX.1 [dev] produced a much higher quality image with better textures and more realistic hands, despite failing the spatial placement of the passenger.

The Halloween Invitation

Text-to-Image

“Vintage gothic Halloween party invitation. Dark parchment poster, spooky border with webs and thorns, central glowing jack-o-lantern, bats, twisted trees, moody night sky. Add elegant gothic title text saying "Halloween Party Invitation", a small scroll banner saying "You are invited to a night of frights", and event details at the bottom: Date: 30.10.2026 Time: 7pm Location: The Arches, NYC Spooky but polished, cinematic lighting, square format.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Features a bold and clear thorn border that fits the gothic theme.
  • + Includes the central glowing jack-o-lantern and bats with vibrant lighting.
  • + Text is legible and uses an appropriate gothic-style font for the theme.
  • Several typos in the text including 'Falloween Ranty' and 'You Tre'.
  • Did not include the requested webs in the border.
  • Redundant time text listed as 'Time: 7pm, 7pm'.

FLUX.1 Krea [dev]

  • + Excellent decorative border with a combination of thorns and detailed spider webs.
  • + Captures the 'vintage parchment' look much more effectively with its aged paper texture.
  • + Better atmospheric composition with more complex tree silhouettes.
  • Major spelling error in the main title reading 'Pasty Halloween' instead of 'Party'.
  • Text in the bottom scroll becomes jumbled and repetitive ('are are', 'nights frights').
  • Date is missing the '10' for the month.

Verdict: Both models struggled significantly with the text requirements, which is the primary metric for an invitation. FLUX.1 [dev] followed the color and lighting cues better but missed the web details and had multiple typos, while FLUX.1 Krea [dev] captured the vintage aesthetic and borders perfectly but failed even more drastically on the spelling of the headline text. FLUX.1 [dev] is the slight winner as its text, despite minor errors, is closer to the intended message and more legible.

Isometric Miniature Diorama Scenes

Text-to-Image

“Create a clear, 45° top-down isometric miniature 3D cartoon scene of Japan's signature dish: sushi, with soft refined textures, realistic PBR materials, gentle lighting, on a small raised diorama base with minimal garnish and plate. Solid light blue background. At top-center: 'JAPAN' in large bold text, 'SUSHI' below it, small flag icon. Perfectly centered, ultra-clean, high-clarity, square format.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent textured rendering of the sushi salmon and rice
  • + Sophisticated miniature aesthetic with depth of field
  • + Accurate 45-degree isometric perspective
  • Text contains a misspelling ('SUSH CATON')
  • Typography is very small and doesn't follow the 'large bold' instruction well

FLUX.1 Krea [dev]

  • + Perfect adherence to text instructions with large, bold, correctly spelled words
  • + Bold and clear graphic composition
  • + Clean lighting and shadows matching the PBR request
  • The sushi texture looks slightly like plastic compared to Model A
  • Composition is very simple with only one piece of sushi

Verdict: Model B (FLUX.1 Krea [dev]) is the winner because it followed all text prompts perfectly, providing the large bold typography requested without spelling errors. While Model A (FLUX.1 [dev]) had a more beautiful 3D texture and miniature feel, its failure to render the requested text size and correct spelling makes it less successful as a prompt-following model.

Adorable Baby Animals in Sunny Meadow

Text-to-Image

“Hyper-photorealistic scene of fluffy baby animals—a golden retriever puppy, tabby kitten, baby bunny, and red fox kit—with big expressive eyes and ultra-detailed soft fur, playfully chasing butterflies and tumbling together in a lush wildflower meadow, warm golden sunrise light with god rays and dew sparkles, joyful wholesome vibe, 8K masterpiece.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent use of warm sunrise lighting and bokeh
  • + Wholesome, storybook-style aesthetic
  • Failed to include a kitten or a fox, opting for four stylized puppy/bunny hybrids instead
  • Style is more 3D-animated/Disney-like than hyper-photorealistic
  • Anatomically wonky paws on the middle characters

FLUX.1 Krea [dev]

  • + Successfully included the golden retriever, tabby kitten, and fox kit
  • + Achieved a high level of realistic fur texture and 'hyper-photorealistic' lighting
  • + Dynamic composition with movement and chasing action
  • Missed the baby bunny in the scene
  • One kitten appears to be floating mid-air with an unnatural posture

Verdict: While both models failed to include all four requested animals, FLUX.1 Krea [dev] followed the prompt's species requirements much better and adhered to the 'hyper-photorealistic' request. FLUX.1 [dev] produced a very cute, high-quality image, but it resembles CGI animation rather than real photography and featured redundant animal types.

Vintage Cafe Logo

Text-to-Image

“Vintage minimalist restaurant logo for "Caffè Florian", retro cloche dome with steam and "Est. 1720" banner, classic typography, warm brown and cream tones, subtle texture on light background, vector emblem style.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Clean vector emblem style that matches the minimalist request.
  • + Accurate inclusion of the 'Est. 1720' text.
  • + Good color palette adherence with warm brown and cream tones.
  • Significant spelling errors in the brand name ('Café Flariláan') and secondary text ('Reseaurant').
  • Includes hallucinated numbers ('11011', '1941') not in the prompt.

FLUX.1 Krea [dev]

  • + Excellent vintage texture and detailed woodcut-style cloche illustration.
  • + Stronger composition with better integration of the cloche and typography.
  • + Includes the 'Est. 1720' text correctly in a stylized frame.
  • Spelling error in the main brand name ('Cafe Flandr'in').
  • Slightly less 'minimalist' than requested due to the high level of detail in the cloche.

Verdict: Both models struggled with the specific spelling of 'Caffè Florian', but FLUX.1 Krea [dev] produced a much more visually appealing and cohesive vintage design. FLUX.1 [dev] followed the minimalist instruction better but suffered from messy text rendering and unnecessary numerical hallucinations.

Apollo 11: Journey to Tranquility

Text-to-Image

“Create a clean, modern vector infographic poster about the Apollo 11 mission. NASA-inspired palette (navy, white, muted red, light gray). Flat-vector style, crisp lines, consistent iconography, subtle gradients only. Steps (stop at landing): 1. Launch (Saturn Vicon) 2. Earth Orbit (Earth + orbit ring icon) 3. Translunar (trajectory arc icon) 4. Lunar Orbit (Moon + orbit ring icon) 5. Descent (lunar module descending icon) 6. Landing (lunar module on the surface icon) Small supporting elements (minimal text): • Crew strip: three silhouette icons with only last names: Armstrong, Aldrin, Collins. • Landing site marker: Moon pin labeled "Tranquility" only. Layout constraints: generous margins, large readable labels, clean background with subtle stars. Vector-only, print-poster look, high resolution.”

FLUX.1 [dev]
FLUX.1 Krea [dev]

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to the requested flat-vector infographic style.
  • + Clean, professional layout with sophisticated muted color palette.
  • + Includes a bottom legend that correctly identifies specific mission phases.
  • The text content for the main steps is largely illegible gibberish.
  • The central rocket icon is somewhat generic compared to a Saturn V.

FLUX.1 Krea [dev]

  • + Clean, readable main title and numbered steps.
  • + Effective use of the 'Tranquility' marker for the landing site.
  • Fails to follow the chronological order requested, jumping from Earth to unrelated astronaut icons.
  • The iconography is inconsistent, featuring four people for a three-man mission and generic ringed planets.
  • Layout feels cluttered and lacks the professional infographic structure of Model A.

Verdict: FLUX.1 [dev] is the clear winner for its superior composition and adherence to the requested infographic style, despite the illegible text. While FLUX.1 Krea [dev] has clearer typeface for secondary labels, its failure to logically follow the mission steps and its inclusion of mismatched icons makes it less effective as an infographic.

Next steps

Explore each model