Unified multimodal model for text-to-image generation, instruction-guided image editing, personalized generation, and virtual try-on
Settled by community votes across 15 shared challenges, with an AI judge weighing in on each.
OmniGen v2
#57 of 62 in Text-to-Image
Wan 2.7 Pro
#38 of 62 in Text-to-Image
Where the votes landed
OmniGen v2
0%
win rate
Ties
0%
Wan 2.7 Pro
0%
win rate
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.”
AI Judge Analysis
OmniGen v2
- + Excellent adherence to lighting instructions with a clear left-to-right gradient
- + High visual clarity and vibrant, clean colors
- + Perfect rendering of the glass cube edges and refractive properties
- − The green plant in the background is less 'partially visible through the glass' and more just behind it
- − The blue sphere has a very plastic, uniform texture
Wan 2.7 Pro
- + Beautiful naturalistic textures on the wooden table and the blue sphere
- + Sophisticated composition with realistic plant interaction through the glass
- + Very high level of detail in the book and window
- − The glass cube has some structural inconsistencies, such as a double vertical line on the front-left face
- − The blue sphere's reflection on the bottom glass pane is slightly misaligned
Verdict: Both models followed the complex spatial prompt perfectly. OmniGen v2 produced a cleaner, more advertising-style image with perfect lighting, while Wan 2.7 Pro achieved a more cinematic and realistic photographic quality with better texture and environmental integration.
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.”
AI Judge Analysis
OmniGen v2
- + Excellent handling of raindrops and atmospheric lighting
- + High-contrast reflections on the wet pavement
- + Clean subject isolation
- − The bike and environment look a bit too clean and 'rendered' rather than candid
- − Missing the requested motion blur on background cars
- − The man is holding the bicycle rather than actively repairing it
Wan 2.7 Pro
- + Features very realistic skin textures and clothing details
- + Captures a genuine 'candid' street photography aesthetic with imperfect framing
- + Better adherence to the 'repairing' action with focused hands
- − The rear wheel of the bicycle has structural inconsistencies where it meets the frame
- − The rain effect is more subtle and less visible than in the other image
- − Lacks the motion blur requested in the prompt
Verdict: Wan 2.7 Pro is the superior choice because it captures the authentic 'candid' and 'no stylization' requirements of the prompt, with a very convincing Japanese street atmosphere. While OmniGen v2 has better atmospheric lighting, it feels overly staged and polished, whereas Wan 2.7 Pro provides the gritty, realistic textures and genuine character interaction requested.
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.”
AI Judge Analysis
OmniGen v2
- + Excellent shallow depth of field with cinematic lighting
- + Intricate engraving details on the armor plates
- + Clear adherence to the beads in hair and dirt on skin requirements
- − The character looks overly pristine and youthful for the 'battle-worn' description
- − The 'dirt' on the face looks more like freckles or makeup than actual grime
Wan 2.7 Pro
- + Perfect interpretation of 'battle-worn' with realistic scars and aged skin
- + Superior texture on leather straps and metal surfaces including tarnish and wear
- + Excellent rendering of braided hair with numerous small beads
- − The composition is slightly less 'close' than Image A, showing more of the torso
- − The background bokeh is a bit busy compared to the smooth blur in Image A
Verdict: While both models followed the prompt well, Wan 2.7 Pro better captured the 'battle-worn' essence through realistic skin textures, convincing scars, and weathered armor. OmniGen v2 produced a beautiful, clean image, but the character felt too much like a model in a costume rather than a veteran paladin.
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.”
AI Judge Analysis
OmniGen v2
- + Strong vibrant color accents
- + Modern abstract graphic style
- − Numerous spelling errors in headings
- − Text is largely illegible placeholder scribbles
- − Poor overall layout balance for a functional menu
Wan 2.7 Pro
- + Exceptional text legibility and plausible spelling
- + Professional, realistic layout with pricing and descriptions
- + High-quality food photography that looks appetizing and consistent
- − Slightly less bold/experimental font than requested
Verdict: Wan 2.7 Pro is significantly better, producing a highly realistic and functional menu design with legible text, coherent food items, and a professional layout. OmniGen v2 fails on basic text generation and suffers from numerous spelling errors and messy layout organization.
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.”
AI Judge Analysis
OmniGen v2
- + Clean, professional graphic design layout
- + Vibrant colors and high-quality textures on the bun and patty
- − Fails the core requirement of an 'exploded' burger, showing a fully assembled one instead
- − Text is partially cut off on the left side
- − Missing the Euro symbol (€) in the price starburst
Wan 2.7 Pro
- + Follows the 'exploded' view prompt perfectly with all components suspended in mid-air
- + Excellent text integration, including the specific fiery effect and currency symbol
- + Dynamic composition with embers, smoke, and motion streaks
- − The 'MAGIC BURGER' text has small fire icons on top that look a bit repetitive
- − The tomato slice texture is slightly less realistic than other components
Verdict: Wan 2.7 Pro followed the instructions much more accurately, successfully creating the complex 'exploded burger' layout and including all requested text elements and currency symbols. OmniGen v2 failed the primary compositional request for an exploded view, providing a standard assembled burger and cutting off part of the secondary text.
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.”
AI Judge Analysis
OmniGen v2
- + Strong chalk texture and smudge effects on the blackboard surface.
- + Correct date and year provided in the prompt.
- − Significant spelling errors throughout the menu items including 'Specals' and 'Lemont'.
- − Text layout is extremely messy with overlapping lines and nonsensical words.
- − Failed to provide the requested 'elegant cursive' for the title.
Wan 2.7 Pro
- + Perfect text rendering with zero spelling errors across all requested items.
- + Excellent layout and composition, fitting the 'cozy café' atmosphere with a realistic background.
- + Successful interpretation of 'elegant cursive' for the header text.
- − The handwriting style is slightly too consistent, leaning toward a digital chalk font rather than manual handwriting.
- − The chalk texture on the letters themselves is a bit too clean.
Verdict: Wan 2.7 Pro is the clear winner as it followed all textual instructions perfectly, including the complex menu items and the specific date. OmniGen v2 struggled significantly with spelling, legibility, and layout, producing a jumbled and incoherent board.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
OmniGen v2
- + Clean, vibrant colors with a cinematic lighting effect
- + High-contrast rendering of the astronaut suit
- − Failed to follow the spatial reasoning instruction 'horse on top'
- − Anatomical issues with the horse's legs appearing to merge or disappear into the background
Wan 2.7 Pro
- + Better background detail with more celestial bodies and a planetary horizon
- + Higher realism in the textures of the horse and astronaut gear
- − Failed to follow the spatial reasoning instruction 'horse on top'
- − Anatomical errors in the horse's front legs and hoof orientation
Verdict: Both OmniGen v2 and Wan 2.7 Pro completely ignored the specific, unconventional request for the 'horse on top' (a horse riding an astronaut), instead generating the clichéd image of an astronaut riding a horse. Wan 2.7 Pro is slightly better due to its more complex composition and realistic textures, although both models struggle with horse anatomy in a zero-gravity context.
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.”
AI Judge Analysis
OmniGen v2
- + The capybara's head has high-quality fur texture.
- + The overall lighting and colors are vibrant and clean.
- − The driver has human hands instead of capybara paws.
- − The businesswoman is sitting in the passenger seat next to the driver instead of the back seat.
- − The capybara's head is not properly attached to the human-looking neck/shoulders.
Wan 2.7 Pro
- + Successfully rendered animal-like paws on the steering wheel.
- + The businesswoman is correctly placed in the back seat as requested.
- + The street background and taxi lighting look highly realistic.
- − The passenger's hands are mangled and fused with the phone.
- − The passenger's scale seems slightly off compared to the interior.
Verdict: Wan 2.7 Pro followed the complex spatial instructions much better than OmniGen v2, correctly placing the passenger in the back seat and providing the capybara with animal paws. While OmniGen v2 has cleaner textures, it failed significantly on prompt adherence by placing the woman in the front seat and giving the capybara human hands.
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.”
AI Judge Analysis
OmniGen v2
- + Successfully renders the large title and central jack-o-lantern.
- + Captures the requested moody night sky and cinematic lighting well.
- − Text rendering is poor with numerous spelling errors and garbled characters in the banner and details.
- − The composition feels a bit like clip-art rather than a polished vintage poster.
Wan 2.7 Pro
- + Excellent text rendering with near-perfect accuracy for the title, banner, and event details.
- + Highly detailed and cohesive composition that effectively incorporates webs, thorns, bats, and lanterns in a vintage style.
- + Creative additions like the cauldron and gravestone enhance the gothic theme without cluttering.
- − The jack-o-lantern's glow is a bit more illustrative and less cinematic compared to Model A.
Verdict: Wan 2.7 Pro is the clear winner as it followed every instruction, including the complex text requirements, with high precision and artistic flair. While OmniGen v2 captured the mood well, its failure to render legible text and its simpler graphic style make it less effective for a party invitation.
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.”
AI Judge Analysis
OmniGen v2
- + Excellent 3D text rendering with shadows that match the lighting.
- + Strong adherence to the 'diorama base' requirement with its tiered pedestal.
- + Clean, saturated colors that pop against the light blue background.
- − The flag icon is generic and does not represent the Japanese flag.
- − The sushi design is a bit repetitive, showing two very similar rolls/pieces.
Wan 2.7 Pro
- + Features a correct Japanese flag icon as requested.
- + Higher variety and complexity in the sushi models (nigiri, maki, gunkan).
- + Excellent material rendering, especially on the roe and ceramic plate.
- − The text is 2D and lacks the 3D isometric integration seen in the other model.
- − The background contains some minor debris/artifacts in the corners.
Verdict: Both models followed the prompt closely, but Wan 2.7 Pro is the superior choice for its accurate flag icon and high-quality variety of sushi types. While OmniGen v2 handled the 3D text shadows better and adhered more strictly to the 'diorama base' concept, Wan 2.7 Pro's overall visual fidelity and material textures are more refined and appealing.
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.”
AI Judge Analysis
OmniGen v2
- + Successfully captures a warm, vibrant sunrise color palette.
- + Displays very clear god rays and soft lighting.
- − Failed the prompt by missing the bunny entirely.
- − Art style is heavily stylized and cartoonish rather than 'hyper-photorealistic'.
- − The animals are static and sitting rather than 'playfully chasing' or 'tumbling'.
Wan 2.7 Pro
- + Includes all four requested animals: puppy, kitten, bunny, and fox kit.
- + Achieves a high level of photorealism in fur textures and environment.
- + Dynamic composition correctly depicts the animals tumbling and chasing butterflies.
- − The fox's front paw has some anatomical blending issues with the surrounding grass/cat.
Verdict: Wan 2.7 Pro followed the complex prompt instructions perfectly, including all four specific animals in a dynamic, photorealistic scene. In contrast, OmniGen v2 missed the bunny, adopted a 3D-animation style instead of realism, and failed to capture the requested action of the animals playing together.
Studio Ghibli Anime Style
Editing“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”
AI Judge Analysis
OmniGen v2
- + Successfully translates the scene into a flat cell-shaded anime style
- + Preserves the composition and poses of the iconic meme perfectly
- + Replaces the background with elements that feel more like a Ghibli neighborhood
- − The facial expressions are too generic and lose the specific emotion (jealousy/surprise) of the original
- − The art style leans more towards modern generic anime than the specific hand-painted Ghibli aesthetic
Wan 2.7 Pro
- + Excellent hand-painted watercolor texture that aligns with Ghibli background art
- + Retains the exact facial expressions and likenesses from the source photo
- + Maintains a high level of source preservation while applying the requested filter
- − The characters retain a photorealistic structure under the texture rather than being redrawn as illustrations
- − The lighting feels a bit washed out compared to the 'warm' mood requested
Verdict: OmniGen v2 provided a complete stylistic overhaul, recreating the scene with traditional anime line art and simplified colors, which fits the 'illustration' part of the prompt well but loses the nuanced facial expressions. Wan 2.7 Pro acted more like a high-quality watercolor filter, preserving the identities and emotions of the original photo perfectly while capturing the 'hand-painted' texture of Ghibli backgrounds. Wan 2.7 Pro is the likely winner as it balances prompt adherence with excellent preservation of the source image's character dynamics.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
OmniGen v2
- + Successfully added wind-blown hair effect
- + Includes floating leaves as requested
- − Significantly alters the woman's facial features compared to the source
- − Leaves appear digital and flat rather than naturally integrated
- − Image looks over-saturated and less photographic than the source
Wan 2.7 Pro
- + Excellent preservation of the woman's facial identity and the dog's appearance
- + Highly natural hair movement and realistic leaf scattering
- + Maintains the lighting and texture of the original photograph
- − Some minor clipping artifacts where hair strands meet the background
Verdict: Wan 2.7 Pro is the clear winner as it perfectly captured the 'dynamic motion' prompt while maintaining the original subject's identity. OmniGen v2 successfully followed the instructions but changed the woman's face so significantly that it no longer feels like the same person from the source image.
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.”
AI Judge Analysis
OmniGen v2
- + Strong minimalist aesthetic with clean vector lines
- + Excellent interpretation of 'warm brown and cream tones'
- + Bold and balanced composition
- − Misspelled the primary brand name as 'CAFFFLORIN'
- − Simplified the steam effect to a single, somewhat abstract line
Wan 2.7 Pro
- + Accurate spelling of 'Caffè' including the accent mark
- + Beautifully detailed etch-style cloche and laurel frame
- + Successfully includes additional relevant details like 'Venezia Italia'
- − Misspelled Florian as 'Florion'
- − Included distracting floating objects in the periphery of the image like coffee beans and cinnamon
- − Slightly less 'minimalist' than the prompt requested
Verdict: OmniGen v2 produced a much cleaner, more professional vector logo that hits the 'minimalist' requirement perfectly, but suffers from a significant spelling error. Wan 2.7 Pro offers a richer, more classic design with better typography for 'Caffè', but it includes messy peripheral elements and also misspells the name of the restaurant. OmniGen v2 is the likely winner for better adhering to the vector emblem style and minimalist aesthetic despite its typo.
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.”
AI Judge Analysis
OmniGen v2
- + Clean vector aesthetic
- + Followed the color palette requirements well
- − Missed the mission number (Apolo 17 instead of Apollo 11)
- − Significant spelling errors and garbled text
- − Failed to include all 6 requested steps in a logical sequence
Wan 2.7 Pro
- + Perfectly followed the sequence of the 6 requested steps
- + Exceptional text rendering and spelling accuracy
- + Professional infographic layout with high information density
- + Clearly adhered to the theme and mission details
- − Minor typo in 'Descent' (rendered as DESCRIPT)
- − Text is very small and may be difficult to read in some areas
Verdict: Wan 2.7 Pro is the clear winner as it successfully incorporated all 6 requested stages of the mission with highly accurate icons and mostly perfect text. OmniGen v2 failed on nearly every instruction, getting the mission name wrong and providing illegible text and confusing iconography.
Explore each model
Alibaba's Wan 2.7 Pro image generation and editing model with higher-quality outputs and support for 4K image generation