Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [klein] 4B Black Forest Labs OmniGen v2 VectorSpaceLab

Settled by community votes across 15 shared challenges, with an AI judge weighing in on each.

FLUX.2 [klein] 4B

22.3 arena score

#32 of 62 in Text-to-Image

Skill signature · Text-to-Image

OmniGen v2

16.9 arena score

#55 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [klein] 4B

0%

win rate

Ties

0%

OmniGen v2

0%

win rate

Shared challenges 15

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photorealism with realistic dust and micro-scratches on the glass.
  • + Correctly renders the cube as hollow with a solid base.
  • + Uses a shallow depth of field that feels more natural for professional photography.
  • The plant visibility through the glass is slightly subtle due to the lighting angle.

OmniGen v2

  • + Strong colors and very clean, vibrant textures.
  • + The red book and blue sphere have high visual impact.
  • The cube geometry is physically impossible, with a floating top edge and a sphere that appears to be floating/intersecting the glass floor.
  • The lighting is flat and lacks the realistic window-light directionality requested.

Verdict: FLUX.2 [klein] 4B produces a significantly more realistic and physically coherent image, capturing the fine details of light, glass texture, and depth of field. OmniGen v2 succeeds in placing the correct objects, but the cube's structure is nonsensical and the rendering looks more like CGI than a photograph.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photographic realism with tactile textures on the jacket and face.
  • + Includes subtle rain streaks and atmospheric lens flare that feel authentic.
  • + Successful 'imperfect' framing and candid composition requested in the prompt.

OmniGen v2

  • + Strong reflections on the wet pavement.
  • + Clear adherence to the 'red bicycle' requirement with vibrant color.
  • The bike geometry is nonsensical, with a frame that doesn't connect logically.
  • The image has a smooth, digital 'AI look' that contradicts the 'no stylization' request.
  • Lacks the requested motion blur from passing cars.

Verdict: FLUX.2 [klein] 4B followed the prompt much more effectively, producing a gritty, realistic photograph with natural skin textures and the requested motion blur in the background. OmniGen v2 struggled with the technical aspects of the prompt, producing a very clean, digital-looking image with significant anatomical and mechanical errors in the bicycle structure.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent depiction of ornate, battle-worn engraved plate armor with dents and scratches.
  • + Highly detailed texture on the leather cross-body strap and cloth neckpiece.
  • + Perfect adherence to 'hair braided with small beads' and 'lifelike eyes'.
  • The torch flames in the background look a bit artificial and detached from the environment.

OmniGen v2

  • + Beautiful cinematic lighting and depth of field with a strong bokeh effect.
  • + Clear and consistent beads in the hair braids.
  • + Very clean facial features and realistic texture in the leather straps.
  • The armor looks very clean and polished, missing the requested 'battle-worn' aesthetic.
  • The dirt on the skin looks more like decorative dots than authentic splatter or grime.

Verdict: FLUX.2 [klein] 4B followed the prompt much more accurately, particularly regarding the 'battle-worn' description and the intricate engraving on the armor. While OmniGen v2 produced an aesthetically pleasing and cinematic image, it ignored the requests for scars and weathered equipment, making the paladin look too pristine for the context.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + High-quality, appetizing food photography with consistent lighting.
  • + Clean and modern professional layout that feels like a real commercial design.
  • + Good typography hierarchy with clear section headings.
  • Nonsense text and significant spelling errors in headings like 'MDAINS'.
  • Some food items bleed together in the grid without sufficient margins.

OmniGen v2

  • + Attempts a multi-page/spread layout which adds to the professional feel.
  • + Uses vibrant color blocking behind food photos as requested for 'vibrant accents'.
  • + Large header text is relatively legible despite misspellings.
  • Severe spelling errors even in large headers ('PIZZZZAN', 'APPTETIZES').
  • Lower resolution in the body text compared to FLUX.
  • The layout is less cohesive, with inconsistent alignment across the grid.

Verdict: FLUX.2 [klein] 4B produces a much more professional and realistic menu design with superior food photography that looks commercially viable. While both models struggled with text rendering, FLUX.2 closer adhered to the 'modern minimalist' aesthetic and provided a more balanced grid layout, whereas OmniGen v2 had more erratic spelling and less clear visual hierarchy.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photorealistic texture on the burger bun and patties
  • + Accurately rendered the secondary text and price tag
  • + Dynamic background with realistic embers
  • Major text failure on the primary title 'Magic Burger'
  • Failed to create an 'exploded' view where components are suspended separately

OmniGen v2

  • + Perfect text rendering for 'Magic Burger'
  • + Bright, illustrative aesthetic suitable for commercial advertising
  • + Includes all requested text elements in the correct layout
  • Failed to provide an 'exploded' view of the burger components
  • Missing the Euro symbol (€) in the price starburst
  • Visual style is more cartoonish/3D render than 'photorealistic'

Verdict: Both models failed the specific 'exploded' composition instruction, showing a standard stacked burger instead. OmniGen v2 is the winner because it successfully rendered the complex text and title accurately, whereas FLUX.2 [klein] 4B completely garbled the name 'Magic Burger'.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent chalk texture with realistic smudges and dust on the board.
  • + Strong adherence to the requested cursive style for the menu items.
  • + Maintains a consistent layout that looks like a real cafe board.
  • Several spelling errors like 'Truffel', 'Musheram', and 'Ootrpous'.
  • The title has an extra 'S' and a space in 'TPECIALS'.

OmniGen v2

  • + Text is highly legible and follows a clear hierarchy.
  • + The board and frame have a clean, modern aesthetic.
  • Significant text corruption and overlapping characters in the middle section.
  • Failed to provide the elegant cursive style requested for the title.
  • The text looks more like a digital font overlay than natural chalk handwriting.

Verdict: FLUX.2 [klein] 4B is the clear winner because it captures the authentic 'chalk' look and feel requested, including realistic textures and handwriting styles, despite some spelling errors. OmniGen v2 struggled significantly with text coherence, resulting in messy overlaps and a style that felt closer to a digital font than natural chalk.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent cinematic lighting and realistic textures on the horse's coat and spacesuit.
  • + Dynamic composition with a sense of motion over the Earth's atmosphere.
  • + High level of detail in the background starfield and planetary curvature.
  • The astronaut's face inside the visor looks slightly distorted.
  • Failed the specific spatial instruction for the horse to be on top of the astronaut.

OmniGen v2

  • + Clean, vibrant colors with a clear distinction between the subject and background.
  • + Simple and iconic character design that reads clearly at a glance.
  • Failed the specific spatial instruction for the horse to be on top of the astronaut.
  • Lower detail level compared to Model A, appearing more like a 3D render than a cinematic image.
  • The two moons in the background are repetitive and lack texture.

Verdict: Both FLUX.2 [klein] 4B and OmniGen v2 failed the negative constraint to have the horse on top of the astronaut, both defaulting to the standard horse-riding pose. FLUX.2 [klein] 4B is the superior image due to its much higher resolution, cinematic texture, and sophisticated lighting, whereas OmniGen v2 feels like a simpler, less detailed digital illustration.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent adherence to the capybara's physical posture with paws on the wheel
  • + High level of realism in texturing of both fur and clothing
  • + Accurately places the passenger in the back seat as requested

OmniGen v2

  • + Clean, vibrant colors in the taxi exterior lighting
  • + Sharp focus on the capybara's facial features
  • Failed the spatial request by putting the passenger in the front seat
  • Human hands are incorrectly attached to the capybara driver
  • Anatomical errors in the passenger's hands and the car's interior layout

Verdict: FLUX.2 [klein] 4B followed the prompt instructions perfectly, placing the capybara and passenger in their correct seats with realistic interactions. OmniGen v2 failed significantly by merging human hands onto the capybara and placing the passenger in the front seat, which contradicted the prompt.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent atmospheric lighting and cinematic quality
  • + Highly detailed background with realistic webs, thorns, and twisted trees
  • + Composed with a strong central focus and spooky aesthetic
  • Numerous spelling errors in the main title and event details
  • Date is missing a digit (206 instead of 2026)

OmniGen v2

  • + Perfectly accurate main title text
  • + Correctly identifies the requested year (2026)
  • + Vibrant colors with a clean digital illustration style
  • Text layout at the bottom is cluttered and repetitive
  • Banner text is unreadable gibberish
  • The 'parchment' effect is a simple border rather than an integrated texture

Verdict: FLUX.2 [klein] 4B produces a much more atmospheric and visually impressive 'vintage gothic' piece, but it fails significantly on text legibility and spelling. OmniGen v2 succeeds in rendering the main title correctly, but the overall design feels less polished and the secondary text becomes a cluttered mess. FLUX.2 is preferred for an artistic base, while OmniGen v2 is slightly better for basic text adherence, but neither fully mastered the complex prompt. FLUX.2's higher visual quality and superior interpretation of the 'parchment' and 'gothic' mood make it the stronger overall image despite the typos.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent texture on the salmon and rice, mimicking a realistic miniature model.
  • + Very sharp, high-resolution rendering of the plate and garnish.
  • + Successfully captures a 45° isometric perspective.
  • Failed to spell 'SUSHI' correctly, omitting the final 'I'.
  • The flag icon is incorrect, resembling the flag of Austria rather than Japan.

OmniGen v2

  • + Accurately spells both 'JAPAN' and 'SUSHI'.
  • + Perfectly interprets the 'small raised diorama base' request.
  • + Bold, clean 3D cartoon aesthetic consistent with the prompt.
  • The flag icon is a generic red/yellow banner rather than the Japanese flag.
  • The sushi roll interior has slightly messy, distorted details.

Verdict: OmniGen v2 is the preferred result because it successfully followed the complex instruction for a raised diorama base and spelled all text correctly. While FLUX.2 [klein] 4B achieved more realistic textures, its failure to complete the word 'SUSHI' and its choice of an incorrect flag make it less accurate to the prompt.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent photorealistic texture on the fur and individual flower petals.
  • + Features dynamic lighting with god rays and dew-like sparkling particles.
  • + Good character count including the puppy, fox, and two kittens.
  • Failed to include the baby bunny as a distinct animal.
  • The puppy has a slightly distorted fifth leg/phantom limb area near the bottom kitten.

OmniGen v2

  • + Successfully captured the 'big expressive eyes' and joyful, cute aesthetic.
  • + Cheerful, vibrant color palette with clear butterfly elements.
  • Has a very artificial, 3D-render or cartoonish look despite the request for photorealism.
  • Missing two of the requested animals (bunny and fox kit/kitten distinction).
  • Low-resolution appearance with blurring and flat textures.

Verdict: FLUX.2 [klein] 4B produces a much more sophisticated and photorealistic image that captures the atmosphere and '8K masterpiece' quality requested, although it missed the rabbit. OmniGen v2 failed the prompt's stylistic requirement for hyper-photorealism, instead delivering a cartoonish, lower-quality 3D render with fewer animals.

Studio Ghibli Anime Style

Editing
Edit instruction

“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”

Source
FLUX.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent adherence to the 'soft pastel' and 'hand-painted textures' prompt requirements.
  • + Preserves the specific light-to-dark plaid pattern of the shirt very accurately.
  • + Captures the 'dreamy' and 'nostalgic' mood through watercolor-style rendering.
  • The facial expressions are softened too much, losing the specific 'shocked' look of the girlfriend.

OmniGen v2

  • + Follows the Ghibli character design style well with bold lines and simplified features.
  • + Maintains a closer color match to the original red dress.
  • + Preserves the composition and background elements like the red bus effectively.
  • Lacks the requested 'hand-painted textures' and 'soft pastel' colors, appearing more like digital vector art.
  • The lighting is flat rather than 'gentle' or 'dreamy' as requested.

Verdict: FLUX.2 [klein] 4B is the winner because it much more effectively captured the specific aesthetic instructions of the prompt, including the hand-painted texture and soft pastel color palette. OmniGen v2 produced a generic anime style that ignored the textural and lighting requests, even though it maintained better character silhouettes.

Golden Hour Stroll

Image Editing
Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

Before After
FLUX.2 [klein] 4B
Before After
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent preservation of the original subjects and background.
  • + Realistic hair motion that looks natural and windswept.
  • + High-quality, detailed leaves that blend well with the scene's lighting.
  • Some leaves in the foreground look slightly flat against the subject.

OmniGen v2

  • + Successful addition of hair motion.
  • + Good preservation of the subject's identity and posture.
  • The added leaves look like cartoonish orange icons over the image.
  • Significant loss of texture and detail across the background and grass.
  • The colors are overly saturated compared to the original, losing the realistic feel.

Verdict: FLUX.2 [klein] 4B followed the instructions perfectly while keeping the integrity of the original photo intact, especially the skin textures and background details. OmniGen v2 added the requested motion but significantly degraded the image quality, resulting in a softened, cartoonish look with unrealistic floating leaves.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Excellent typography style that matches the vintage aesthetic.
  • + Beautiful paper-like subtle texture on the background.
  • + Clean vector emblem design with great shading on the cloche.
  • Misspelled the name as 'CAFFÈ FLAXTION'.
  • Redundant 'Est. 1720' text below the banner.

OmniGen v2

  • + Successfully incorporated the banner and establishing date clearly.
  • + Achieved the requested minimalist vector vibe.
  • + Includes the steam element as requested.
  • Misspelled the name as 'CAFFFLORIN'.
  • Composition is a bit bottom-heavy with the thick banner and date.
  • Failed to include a subtle texture on the background.

Verdict: Both models struggled with the specific spelling of 'Caffè Florian', with FLUX.2 [klein] 4B inventing 'Flaxtion' and OmniGen v2 mashing the words together into 'Caffflorin'. However, FLUX.2 [klein] 4B produced a much more professional-looking vector design with superior typography and texture, making it the preferred choice for a logo concept despite the spelling error.

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.2 [klein] 4B
OmniGen v2

AI Judge Analysis

FLUX.2 [klein] 4B

  • + Successfully included complex icons for the Saturn V and Lunar Module.
  • + Captures the NASA-inspired color palette and flat vector style effectively.
  • + Includes silhouetted crew members as supporting details.
  • Text rendering is very poor with several gibberish words.
  • The layout is a bit cluttered and lacks the logical flow of a 6-step process.

OmniGen v2

  • + Stronger infographic layout with a clear grid and headers.
  • + Minimalist, clean iconography that aligns well with the 'flat-vector' request.
  • + Superior font legibility, even though spelling errors are present.
  • Failed to include a Saturn V icon or a Lunar Module, using generic circular icons instead.
  • Misidentified the mission as 'Apolo 17' instead of Apollo 11.
  • Only shows 3 main sections instead of the requested 6 steps.

Verdict: FLUX.2 [klein] 4B follows the content of the prompt much better, providing specific mission assets like the Saturn V and Lunar Module, though it struggles significantly with text and layout. OmniGen v2 creates a much cleaner and more professional-looking 'graphic design' piece, but fails to represent the actual components of the Apollo 11 mission requested. FLUX.2 is the winner for actually attempting the specific technical icons, despite the messy typography.

Next steps

Explore each model