Head to head
Esc

Models · slot A

to navigate to pick

OmniGen v2 VectorSpaceLab Wan 2.7 Alibaba

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

OmniGen v2

16.8 arena score

#57 of 62 in Text-to-Image

Skill signature · Text-to-Image

Wan 2.7

20.5 arena score

#39 of 62 in Text-to-Image

Vote tally

Where the votes landed

OmniGen v2

0%

win rate

Ties

0%

Wan 2.7

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Excellent clean aesthetic with high-quality rendering of the glass and sphere.
  • + Light refraction and reflections on the sphere and table are highly realistic.
  • + The plant is visible through the glass exactly as requested.
  • The glass cube appears to have an open top or a slightly physically impossible connection with the book.
  • The composition is a bit tight, cutting off the bottom of the table reflection.

Wan 2.7

  • + Features a highly realistic rustic wooden table with convincing texture.
  • + The book has a more natural, aged texture with visible spine details.
  • + Good spatial awareness with internal reflections of the sphere on the glass walls.
  • The glass cube has strange internal vertical bars or seams that don't match a simple cube.
  • The blue sphere has an odd, non-spherical textured surface.
  • The glass panels look more like a display case with a base rather than a solid glass cube.

Verdict: OmniGen v2 produces a much cleaner and more aesthetically pleasing image with superior lighting and refraction effects that perfectly follow the prompt. While Wan 2.7 has better textures for the wood and book, it fails on the 'glass cube' geometry by including distracting internal seams and a misshapen sphere.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Excellent clear reflections on the wet pavement.
  • + Bold colors and clean subject-background separation.
  • The image looks overly clean and CG-like, failing the 'no stylization' and 'candid' requirements.
  • The bicycle geometry is broken, with the frame passing through the man's leg.
  • Missing requested motion blur from passing cars.

Wan 2.7

  • + Strong adherence to the 'candid street photo' aesthetic with realistic lighting and imperfections.
  • + Accurately captures the 'light rain' and 'natural skin texture' for a non-stylized look.
  • + Composition feels much more authentic to a 50mm lens on a busy street.
  • The bicycle frame geometry is slightly warped near the handlebars.
  • The background characters are a bit blurry, though this fits the shallow depth of field.

Verdict: Wan 2.7 successfully captured the gritty, realistic, and candid nature of the prompt, looking like a genuine photograph taken in Japan. In contrast, OmniGen v2 produced a highly stylized, clean rendering that felt like digital art and contained significant anatomical/object clipping errors, such as the bike frame merging with the man's body.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Excellent shallow depth of field and bokeh effect
  • + Intricate engraving on the gold-toned armor plates
  • + Soft, painterly lighting that creates a strong mood
  • The character looks too clean and polished for a 'battle-worn' description
  • Dirt marks look like flat dots rather than natural grime or texture

Wan 2.7

  • + Perfect adherence to the 'battle-worn' prompt with realistic scars and skin texture
  • + Highly detailed leather straps, buckles, and metallic pitting
  • + Authentic braided hair with complex beadwork
  • The lighting on the face is slightly flat compared to the dramatic background source
  • Some overlapping braid elements merge slightly into the armor

Verdict: OmniGen v2 produces a more aesthetically stylized image with beautiful lighting, but it fails to capture the 'battle-worn' and 'scarred' nature of the prompt. Wan 2.1 delivers exceptional realism and detail, accurately portraying a weathered warrior with high-fidelity textures on the armor, leather, and skin.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Features a grid layout with vibrant color-block accents.
  • + Includes the requested pizza and appetizer categories.
  • + High-quality, realistic individual food photos.
  • Numerous spelling errors in headings like 'RESTAURATED MENTS' and 'APPETIZES'.
  • Layout feels a bit cluttered and unbalanced between the two pages.
  • Section titles are repetitive and nonsensical in places.

Wan 2.7

  • + Excellent typography with legible titles and prices.
  • + Professional, clean layout that looks like a ready-to-use menu.
  • + Strong adherence to all prompt elements including grid design and specific sections.
  • Includes unrequested background props like a pen and salt bowl which slightly distracts from the menu itself.
  • Some minor typos in smaller descriptive text.

Verdict: Wan 2.7 is the clear winner as it produces a professional, functional menu layout with legible text, pricing, and a highly organized grid. While OmniGen v2 has vibrant colors, its significant spelling errors and disjointed layout make it less effective as a graphic design piece.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Clean, readable graphic design for the main title.
  • + Vibrant color palette and good lighting on the bun.
  • Failed the primary prompt requirement for an 'exploded' burger with suspended components.
  • Missing the currency symbol (€) in the starburst.
  • Some text is cut off on the left side.

Wan 2.7

  • + Perfectly captured the 'exploded' burger concept with suspended ingredients.
  • + Accurately rendered all requested text including the currency symbol.
  • + Excellent adherence to the fiery aesthetic and dynamic sense of motion.
  • The sesame seeds floating on the left appear a bit disconnected and grainy.
  • The pickles were not explicitly requested, though they fit the theme.

Verdict: Wan 2.7 followed every detail of the prompt, successfully creating a dynamic, exploded view of a burger with all required text and currency symbols. OmniGen v2 failed the core mechanical request of an exploded burger, instead providing a standard stacked burger with poor text placement and missing characters.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Captures a more authentic chalk texture on the surface.
  • + Follows the request for a simple chalkboard frame well.
  • Significant spelling errors throughout, including 'SPECALS', 'Lemont', and illegible scribbles.
  • Very messy layout with text overlapping and inconsistent pricing placement.
  • Handwriting looks like a digital brush rather than a natural slant.

Wan 2.7

  • + Perfect text rendering for all requested menu items and numbers.
  • + Excellent composition with a cozy café background that matches the prompt's atmosphere.
  • + Realistic chalk smudging and texture on the board itself.
  • The handwriting style is very uniform, leaning towards a digital font look rather than varied handwriting.
  • The 'elegant cursive' request for the title was not fully met, as it is a serif-style script.

Verdict: Wan 2.7 is the clear winner as it successfully rendered every word of the complex prompt accurately and legibly, whereas OmniGen v2 struggled with severe spelling errors and layout chaos. Wan 2.7 also provided a high-quality background setting that enhanced the 'cozy café' atmosphere requested.

The Reversed Rodeo

Text-to-Image

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

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Clean, illustrative style with bold colors
  • + Clear rendering of the astronaut's space suit and face shield
  • + Correctly places the astronaut on the horse
  • Fails the specific negative constraint 'horse on top'
  • Anatomical issues with the horse's back legs becoming sheer or missing hoof detail
  • Background lacks the requested 'cinematic' depth, appearing more like a 2D backdrop

Wan 2.7

  • + Highly detailed cinematic background with galaxies and orbital perspective
  • + Realistic textures on the horse's coat and leather saddle
  • + Compositionally superior with dynamic lighting and atmospheric depth
  • Fails the specific negative constraint 'horse on top'
  • Minor clipping issue where the horse's tail meets its body

Verdict: Both models completely failed the semantic nuance of the prompt, which specifically requested the horse be on top of the astronaut. However, Wan 2.7 produced a much higher quality image with superior details, lighting, and cinematic composition compared to the flatter, more artificial look of OmniGen v2.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Features a more photorealistic render of the capybara's fur and face.
  • + The lighting on the interior feels more consistent with a nighttime street scene.
  • The driver's hands are human instead of capybara paws.
  • The human passenger is in the front seat together with the driver, failing the 'back seat' part of the prompt.

Wan 2.7

  • + Correctly renders capybara paws on the steering wheel.
  • + Includes more realistic New York taxi details like the rooftop light.
  • + Successfully places the passenger in the background to imply the back seat area.
  • The fur texture on the capybara's neck looks a bit like artificial bristles rather than soft fur.
  • The human passenger is looking away from her phone rather than at it.

Verdict: OmniGen v2 produces a more polished individual subject, but fails significantly on logic by giving the capybara human hands and placing the passenger in the front seat. Wan 2.7 adheres much better to the prompt's structural requirements, correctly including capybara paws and a background passenger, while creating a more authentic taxi atmosphere.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Excellent typography for the main title
  • + Vibrant, spooky lighting on the jack-o-lantern
  • Serious spelling errors in the date and location details
  • Layout is crowded and the scroll text is garbled
  • Digital graphic look rather than a vintage parchment feel

Wan 2.7

  • + Perfect text rendering for all details including date and location
  • + Highly detailed vintage gothic illustration style with a clear thorn border
  • + Excellent composition with a functional scroll banner
  • The date 'Est. 1847' was not requested in the prompt
  • Slightly less 'cinematic' lighting compared to the center of Image A

Verdict: Wan 2.7 is the clear winner as it successfully rendered every piece of requested text accurately, including the specific date and location. While OmniGen v2 has a striking central graphic, its inability to legibly spell 'The Arches' or 'night of frights' makes it unsuccessful as an invitation. Wan 2.7 also captures the 'vintage gothic' aesthetic much more effectively with its intricate line work and parchment texture.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Excellent typography with clean bold text and 3D shadowing.
  • + Vibrant lighting and highly saturated colors.
  • The flag icon is incorrect, showing a red and yellow design instead of the Japanese flag.
  • The sushi anatomy is slightly nonsensical, merging a nigiri tail with a maki-style roll.

Wan 2.7

  • + Includes a correct Japanese flag icon.
  • + Higher variety of sushi types and better material realism on the fish and rice.
  • + Cleaner, more professional isometric composition.
  • The text layout is slightly tighter and less dynamic than the 3D style in the other image.

Verdict: Wan 2.7 is the superior model for this prompt because it followed the instruction for a specific flag icon correctly and produced a more coherent set of 3D miniature sushi. While OmniGen v2 had great text rendering, its failure to represent the Japanese flag and the odd anatomical blending of the sushi pieces made it less accurate.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Features vibrant, saturated colors and clear backlighting
  • + Includes the requested butterfly elements and golden lighting
  • Fails to include all four animals, missing the bunny entirely
  • Has a very cartoonish, digital illustration style rather than the requested hyper-photorealistic scene
  • Composition is static and lacks the 'tumbling' or 'playful chasing' action requested

Wan 2.7

  • + Successfully includes all four requested animals: puppy, kitten, bunny, and fox kit
  • + Achieves a high level of photorealism with detailed fur textures and realistic anatomy
  • + Captures the action of chasing and tumbling with dynamic posing and great depth of field
  • The fox's facial proportions are slightly narrow
  • A few minor anatomical glitches where the kitten's paw meets the bunny

Verdict: Wan 2.7 is the clear winner as it followed all prompt instructions, including the presence of four distinct animals and a photorealistic style. OmniGen v2 failed the primary prompt requirements by producing a cartoonish illustration and omitting the bunny.

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
OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Strong 'modern anime' aesthetic that aligns with some Ghibli character design traits.
  • + Effective transformation of background elements into clean, stylized scenery.
  • + Good use of warm lighting and bright colors.
  • Completely loses the emotional tension and specific facial expressions of the landmark meme.
  • Changes the composition by moving characters closer together and altering eye contact.

Wan 2.7

  • + Identifies and beautifully applies a 'hand-painted' watercolor texture consistent with Ghibli background art.
  • + Preserves the subject identity and specific facial expressions perfectly.
  • + Maintains the exact composition of the original meme.
  • The character style leans more towards realistic watercolor illustration than the iconic Ghibli character design.

Verdict: OmniGen v2 successfully captures a generic anime look but fails as an edit because it removes the specific emotional context and character poses of the original meme. Wan 2.7 provides a much better edit by applying a high-quality hand-painted texture while preserving the exact composition and expressions, even if it feels slightly less like a character cel than Model A.

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
OmniGen v2
Before After
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Successfully added a wind effect to the hair.
  • + Included floating orange leaves as requested.
  • Significantly altered the original subject's facial features and body proportions.
  • Lost original background details like the specific bridge architecture and texture.
  • Overall image style became much more saturated and 'AI-smoothed' compared to the source.

Wan 2.7

  • + Excellent preservation of the source image's identity, background, and lighting.
  • + Integrated realistic, wind-blown hair that matches the original subject's style.
  • + Added natural-looking falling leaves that blend well with the existing scene.
  • The leash handle area has some minor artifacting where it meets the hand.

Verdict: Wan 2.7 is the clear winner as it perfectly follows the editing instructions while maintaining the integrity of the source image. OmniGen v2 essentially regenerated the entire scene, resulting in a different person and a stylized, less realistic aesthetic that fails the primary goal of source preservation.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Strong minimalist vector style
  • + Clean and simple composition typical of a modern logo
  • + Follows the brown and cream color scheme well
  • Significant spelling error in the main name ('CAFFFLORIN')
  • Lacks the requested 'Caffè' accent mark
  • The steam effect is very basic and abstract

Wan 2.7

  • + Accurate spelling including the 'Caffè' accent mark
  • + Beautifully textured background and sophisticated vintage aesthetic
  • + Detailed cloche dome with better visual interest
  • The main name contains a minor typo ('Florion' instead of 'Florian')
  • Less 'minimalist' than the prompt requested with many decorative elements

Verdict: Wan 2.7 provides a much more visually appealing and professional-looking vintage emblem that captures the 'Caffè Florian' brand identity better, despite a small typo in the last name. OmniGen v2 successfully achieved a more minimalist look, but failed significantly on the primary text spelling and overall artistic quality compared to Wan 2.7.

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.”

OmniGen v2
Wan 2.7

AI Judge Analysis

OmniGen v2

  • + Successfully uses the requested flat-vector style with crisp lines.
  • + Captures the NASA-inspired color palette correctly.
  • Failed the prompt sequence requirements, hallucinating 'Apolo 17' instead of Apollo 11.
  • Text is largely gibberish throughout the infographic.
  • Icons are generic and do not clearly represent the requested mission stages.

Wan 2.7

  • + Excellent adherence to the specific 6-step sequence requested in the prompt.
  • + Superior typography and spelling for a generative model.
  • + Accurately renders all icons including the Saturn V, Earth/Moon orbits, and Lunar Module.
  • Minor spelling error on 'Descent' (spelled 'Descript' or 'Descent' rate as 'Deccent').
  • Small layout artifact where the descent icon is slightly cluttered.

Verdict: Wan 2.7 followed the complex multi-step instructions perfectly, creating a highly legible and accurate vertical infographic with correct iconography for every stage of the mission. OmniGen v2 failed significantly on prompt adherence, producing the wrong mission number, incoherent text, and only three vague steps instead of six.

Next steps

Explore each model