Head to head
Esc

Models · slot A

to navigate to pick

FLUX.1 [dev] Black Forest Labs OmniGen v2 VectorSpaceLab

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

OmniGen v2

16.9 arena score

#56 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 [dev]

0%

win rate

Ties

0%

OmniGen v2

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealistic texture on the red book cover
  • + Sophisticated glass refraction and internal reflections
  • + Perfectly captures the requested soft window lighting
  • The glass structure is slightly non-standard with thick vertical internal panes
  • The blue sphere appears to be floating unnaturally

OmniGen v2

  • + Simple and clean glass cube construction
  • + Good adherence to all spatial requirements in the prompt
  • + Accurate focus and depth of field
  • The red book has a slightly plastic or artificial texture
  • The lighting on the cube is a bit flat compared to the complex light in Model A

Verdict: FLUX.1 [dev] produced a more visually stunning image with impressive textures and complex light behavior, though the internal geometry of the glass is a bit abstract. OmniGen v2 followed the literal spatial instructions very well but resulted in a slightly more 'rendered' look with less realistic textures on the book.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent shallow depth of field and realistic bokeh
  • + Highly detailed skin texture and fabric rendering
  • + Strong adherence to the 'candid street photo' aesthetic with motion blur
  • The scale of the man relative to the bike makes the bike look slightly small
  • The man's right hand has some minor anatomical awkwardness

OmniGen v2

  • + Very clear reflections on the wet pavement
  • + Good central composition of the subject
  • Rain looks like a static filter rather than integrated into the scene
  • Lacks the requested motion blur for the passing cars
  • The man is straddling the bike rather than repairing it

Verdict: FLUX.1 [dev] successfully captured the cinematic and realistic photography style requested, particularly with its natural skin textures and convincing motion blur. OmniGen v2 failed to incorporate the motion blur and the rain effect felt artificial, resulting in a less realistic and more staged appearance. FLUX.1 [dev] is the clear winner for following the specific technical photography instructions and achieving a higher level of photorealism.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent skin texture with realistic pores and freckles.
  • + Intense, lifelike eye detail and lighting.
  • + High-quality metallic reflections on the armor.
  • Failed to include the beads in the braids requested by the prompt.
  • The subject looks more like a modern woman in armor than a battle-worn warrior.
  • Missing the specific engraved floral/ornate patterns mentioned.

OmniGen v2

  • + Accurately included the small beads within the braided hair.
  • + Clearly defined ornate engraving on the plate armor.
  • + Better representation of dirt on the skin and a more cinematic lighting composition.
  • Skin texture is slightly smoother and less realistic than its competitor.
  • The bokeh fire in the background is a bit distracting compared to the requested sparks.

Verdict: OmniGen v2 is the winner as it followed more specific prompt instructions, including the beads in the hair and the ornate engraving on the armor. While FLUX.1 [dev] produced a more lifelike facial texture, it missed several key descriptive elements that OmniGen v2 successfully integrated.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent typography with very clean, readable headers
  • + Highly professional minimalist layout that looks like a real-world design
  • + Consistent and appealing food photography style
  • Missed the 'grid' requirement for the images
  • The food items are positioned somewhat randomly rather than following the section structures

OmniGen v2

  • + Successfully followed the 'grid' layout instruction for food photos
  • + Bold use of color and vibrant accents as requested
  • + Includes more variety in food images
  • Significant spelling errors in large titles (e.g., 'PIZZZZAS', 'RESTAURATED MENTS')
  • The body text is illegible scribbles rather than structured placeholder text
  • The layout feels cluttered and less professional compared to the other model

Verdict: FLUX.1 [dev] produced a much more realistic and professional menu with superior typography and layout, even though it ignored the 'grid' instruction. OmniGen v2 adhered more closely to the grid layout and vibrant colors, but suffered from severe spelling errors and messy details. overall, FLUX.1 [dev] is preferred for its design quality and legibility.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealistic texture on the meat and buns
  • + Creative interpretation of the 'exploded' request with multiple floating layers
  • + Clean and accurate price symbol rendering
  • Completely failed to include the primary 'MAGIC BURGER' title text
  • Failed to include the requested starburst for the price

OmniGen v2

  • + Highly accurate adherence to all requested text elements
  • + Includes the starburst element specifically requested in the prompt
  • + Bold, graphic composition suitable for an advertisement
  • Failed to provide an 'exploded' burger, showing a fully assembled one instead
  • Less photorealistic and more illustrative/cgi-style textures

Verdict: FLUX.1 [dev] produced a much more realistic and visually interesting 'exploded' burger, but it completely ignored the core text requirements of the prompt. OmniGen v2 followed all text and graphic instructions (like the starburst and title) perfectly, but failed to execute the 'exploded' physical layout, presenting a static burger instead. OmniGen v2 is the winner for following the complex advertising requirements more comprehensively, despite the lack of suspension.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent text legibility and spelling accuracy.
  • + Very realistic chalk texture and natural-looking handwriting variations.
  • + High-quality background bokeh creates a convincing café atmosphere.
  • The cursive title is fairly simple rather than 'elegant'.
  • The bottom lines of text have slight artifacts like 'drish' instead of 'dish'.

OmniGen v2

  • + Successfully captured the requested date and main title.
  • + Handwriting has a distinct, varied chalk look.
  • Significant spelling errors and jumbled text throughout the menu items.
  • Poor layout with overlapping text and nonsensical word fragments.
  • Lacks the atmospheric background depth requested in the 'cozy café' prompt.

Verdict: FLUX.1 [dev] is the clear winner as it successfully rendered almost all the requested text with high legibility and realistic chalk textures. OmniGen v2 struggled significantly with the menu items, producing garbled text and numerous spelling errors despite getting the title correct.

The Reversed Rodeo

Text-to-Image

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

FLUX.1 [dev]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent cinematic lighting and atmosphere.
  • + Good anatomical integration of the rider and horse in a zero-gravity pose.
  • + Highly detailed textures on the spacesuit and horse mane.
  • Failed the negative constraint to have the horse on top of the astronaut.
  • Anatomy of the horse's back legs is slightly distorted/elongated.

OmniGen v2

  • + Clean, vibrant colors with a clear celestial background.
  • + Sharp rendering of the astronaut's equipment and patches.
  • Failed the crucial logical constraint to have the horse on top of the astronaut.
  • The composition is very static and lacks the 'cinematic' feel requested.
  • Horse's back legs appear to have multiple joints or odd proportions.

Verdict: Both FLUX.1 [dev] and OmniGen v2 failed the difficult negative constraint to have the 'horse on top.' However, FLUX.1 [dev] produced a much more cinematic and visually impressive image with superior lighting and a more natural-feeling composition for a space setting, whereas OmniGen v2 felt like a static 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.1 [dev]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealistic texture on the capybara's fur.
  • + Accurately places the passenger in the back seat as requested.
  • + Captures the professional, calm expression and specified paw placement on the wheel.

OmniGen v2

  • + Sharp, clear lighting on the subject's face.
  • + Accursately renders a dark jacket and yellow cap as requested.
  • Failed to place the passenger in the back seat; she is in the front passenger seat.
  • Human hands are attached to the capybara's body instead of paws.
  • Strong AI artifacts where the cap meets the capybara's head.

Verdict: FLUX.1 [dev] followed all spatial and anatomical instructions, correctly placing the businesswoman in the back seat and depicting the capybara with paws. OmniGen v2 struggled with the layout, placing the passenger in the front, and failed significantly on the 'paws' instruction by generating human arms for the animal.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent thorns and branch border detail.
  • + Professional cinematic lighting and high-quality illustration style.
  • + Strong adherence to the 'vintage gothic' and 'dark parchment' aesthetic.
  • Significant spelling errors in the title text ('Falloween Rantcy').
  • Duplicate timing information in the footer.
  • Missing the required spider web elements mentioned in the prompt.

OmniGen v2

  • + Successfully included all elements including spider webs and thorns.
  • + Superior text rendering for the main title heading.
  • + Accurate parchment paper texture and border framing.
  • The scrolling banner text is garbled and unreadable.
  • Layout feels a bit cluttered with overlapping text at the bottom.
  • Some fine text details like 'FRIGHTS' contain minor spelling artifacts.

Verdict: OmniGen v2 performed significantly better on the text-heavy requirements, correctly spelling the main title while FLUX.1 [dev] produced incomprehensible gibberish for the 'Halloween' heading. While FLUX.1 [dev] produced a more polished and artistic jack-o-lantern illustration, OmniGen v2 adhered more closely to the full prompt instructions by including the spider web decorative elements.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent soft textures and PBR materials that achieve a realistic miniature look.
  • + Good adherence to the 'gentle lighting' and 'sof refined textures' prompt.
  • + Creative inclusion of miniature cherry blossom elements.
  • Text rendering is hallucinated and inaccurate ('SUSH CATON' instead of 'SUSHI').
  • The 45-degree isometric angle is slightly off, feeling more like a standard perspective shot.

OmniGen v2

  • + Perfect text rendering of 'JAPAN' and 'SUSHI' in large bold font.
  • + Strict adherence to the 45° isometric perspective and diorama base request.
  • + Highly vibrant and clean cartoon aesthetic with a distinct 3D miniature feel.
  • The flag icon is generic/incorrect rather than specifically a Japanese flag.
  • The material appearance is more 'plastic flat' rather than the requested 'realistic PBR' textures.

Verdict: OmniGen v2 followed the layout and text instructions perfectly, providing a clean isometric diorama with accurate labels. FLUX.1 [dev] produced better lighting and more sophisticated material textures but failed significantly on the text and flag accuracy.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Captures the warm golden hour lighting and dew sparkles effectively.
  • + Excellent composition with a natural-feeling depth of field.
  • + The fur textures are very soft and well-integrated into the lighting.
  • Completely failed to include the tabby kitten and substituted for two puppies.
  • The fox kit and bunny are stylized to a point where they look identical in anatomy.
  • Violates the prompt's request for 'photorealistic' by adopting a 3D-animation/Pixar style.

OmniGen v2

  • + Successfully included distinct species (puppy, kitten, fox/bunny hybrid).
  • + Bright, vibrant colors that match the 'joyful vibe' specified.
  • + Better adherence to the 'god rays' request with distinct light beams.
  • Misses one of the four required animals (either the fox or bunny is merged/missing).
  • Anatomy is very cartoonish and far from the requested 'hyper-photorealistic' style.
  • Butterflies look like flat stickers pasted onto the image rather than integrated elements.

Verdict: Both models failed to deliver a truly 'hyper-photorealistic' scene, instead opting for a cutesy, 3D-animated aesthetic. FLUX.1 [dev] produced a much more polished and aesthetically pleasing image with superior lighting, but failed significantly on prompt adherence by duplicating the puppy and omitting the kitten. OmniGen v2 adhered better to the character count and variety but displayed lower technical quality and less sophisticated composition.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Elegant typography style that fits the vintage aesthetic.
  • + Beautifully detailed etching style on the banner and cloche.
  • + Matches the warm brown and cream color scheme perfectly.
  • Significant spelling errors in the brand name ('Flarilaan') and description ('Reseaurant').
  • Includes random numbers (11011, 1941) not requested in the prompt.

OmniGen v2

  • + Very clean, minimalist vector design perfect for a logo.
  • + Better centered composition and layout balance.
  • + Closer to the requested 'Est. 1720' prominence.
  • Spelling error in the brand name ('CAFFFLORIN').
  • The cloche illustration is perhaps a bit too simplistic compared to the vintage request.

Verdict: While both models failed to spell the brand name correctly, OmniGen v2 produced a much more functional logo design with clean vector lines and a professional layout. FLUX.1 [dev] captured the vintage texture and etching style more effectively but cluttered the image with hallucinated numbers and severe typos in the banner text.

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]
OmniGen v2

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to the NASA-inspired color palette and flat vector aesthetic.
  • + Complicated layout that successfully integrates a central launch icon with orbital paths.
  • + Clear consistent iconography and high visual appeal for a poster format.
  • Text consists of gibberish despite the prompt's clear step titles.
  • Includes irrelevant planetary bodies like Saturn and Mars that weren't requested and don't fit the Apollo 11 theme.

OmniGen v2

  • + Captures a very clean, structured infographic layout with clear sections.
  • + Better attempt at legible text, even if misspelled ('APOLO 17', 'TRANQUILITY').
  • + Adheres strictly to the requested red, white, and navy color scheme.
  • Failed the core prompt instruction by labeling it 'Apolo 17' instead of Apollo 11.
  • Iconography is overly simplistic and doesn't clearly reflect the stages specified (e.g., Saturn V, lunar module).

Verdict: FLUX.1 [dev] produced a much more visually sophisticated and artistic vector illustration that feels like a professional poster, though it failed to use the specific text labels requested. OmniGen v2 followed the layout requirements of an infographic more literally but failed the basic subject matter by changing the mission to Apollo 17 and providing very abstract icons. FLUX.1 [dev] is the preferred choice for its superior artistic execution and style adherence.

Next steps

Explore each model