ByteDance's latest image generation model unifying text-to-image and image editing in a single architecture, with improved text rendering and 30-40% faster generation than v4.0
Settled by community votes across 5 shared challenges, with an AI judge weighing in on each.
Seedream 4.5
26.1
arena score
#10 of 44 in Text-to-Image
Skill signature
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
Wan 2.7
19.0
arena score
#34 of 44 in Text-to-Image
Vote tally
Where the votes landed
Seedream 4.5
100.0%
win rate
Ties
0.0%
Wan 2.7
0.0%
win rate
100.0%
0.0% ties
0.0%
Shared challenges
5
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
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.”
Seedream 4.5
Wan 2.7
100% wins
0% ties
0% wins
AI Judge Analysis
Seedream 4.5
- + Features realistic chalk texture with varying opacity and dust particles.
- + Successfully renders handwritten text with natural variations in letter size and slant.
- + Accurately completes truncated items from the prompt like 'Brown Butter Chocolate Chip Cookies'.
- − Repeats the word 'Risotto' and the price '$24' unnecessarily on two lines.
- − Text slant and spacing are slightly messy, though this fits the 'handwritten' request.
Wan 2.7
- + Text is perfectly centered and highly legible.
- + Includes all requested menu items with no spelling errors.
- + Clean, professional composition with nice lighting and background elements.
- − Failed the negative constraint; the text looks like a clean digital font rather than authentic chalk handwriting.
- − The 'handwriting' is too uniform, lacking the requested slants and natural variations.
Verdict: Seedream 4.5 much better captures the intended 'chalkboard' aesthetic, featuring authentic textures, smudges, and imperfect handwriting that truly looks hand-drawn. While Wan 2.7 has cleaner layout and spelling, it relies on a digital-looking font that ignores the specific prompt instructions for natural variations and a non-digital appearance.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
Seedream 4.5
Wan 2.7
AI Judge Analysis
Seedream 4.5
- + Successfully followed the difficult positional constraint with the horse on top of the astronaut.
- + Dynamic and cinematic lighting with vibrant nebula colors.
- + Excellent anatomical detail on the horse's head and neck.
- − The astronaut's lower body merging into the horse's back is a bit messy.
- − The perspective of the astronaut's legs is slightly distorted.
Wan 2.7
- + Clean, high-resolution rendering of the astronaut and planets.
- + Stable composition with clear celestial objects.
- + Good lighting consistency between the subjects and the orbital background.
- − Completely failed the negative constraint, depicting the astronaut riding the horse.
- − Nonsensical repetition of Earth scattered in the background.
- − Lacks the 'surreal' quality requested, looking like a standard collage.
Verdict: Seedream 4.5 is the clear winner because it correctly interpreted the specific request for the horse to be on top of the astronaut, creating a truly surreal image. Wan 2.7 ignored the positional instruction and generated a standard astronaut riding a horse, while also featuring odd repeated Earths that diminished the visual quality.
Outfit Transfer Challenge
EditingEdit instruction
“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”
Source
Seedream 4.5
Wan 2.7
AI Judge Analysis
Seedream 4.5
- + Excellent adherence to the specific clothing items from Image 2
- + Maintains the exact face and hair of the person from Image 1
- + Includes the specific accessories like the ring and gold watch
- − Fails to realistically layer the shirt over the person's torso, resulting in a strange cutout of the skin
- − The pose is slightly altered from the original image
- − Poor proportions on the hands
Wan 2.7
- + Successfully adapts a new pose that feels natural in the environment
- + Maintains the likeness/skin patterns of the base person well
- + High visual quality and resolution
- − Completely ignored the clothing in Image 2, creating a generic 'elaborate' outfit instead
- − Altered the layout and composition of the original scene significantly
- − The hands and fingers have anatomical errors
Verdict: Seedream 4.5 followed the complex instructions much better by using the actual coat, scarf, and jeans from Image 2, even though the internal rendering of the shirt and skin is messy. Wan 2.7 failed the primary task of using the 'exact' outfit, instead generating a random decorative suit that was never present in the source images. Seedream 4.5 is the preferred choice for following the specific edit constraints despite its technical artifacts.
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.”
Seedream 4.5
Wan 2.7
100% wins
0% ties
0% wins
AI Judge Analysis
Seedream 4.5
- + Excellent adherence to the 'inside the taxi' perspective.
- + High-quality fur texture and realistic passenger expression.
- + Clean, readable text on the taxi cap.
- − The perspective makes the capybara look giant compared to the passenger.
- − The lighting on the front of the capybara is slightly flat.
Wan 2.7
- + Natural composition from an outside-looking-in perspective.
- + Very detailed capybara hands/paws on the wheel.
- + Good lighting and integration with the city backdrop.
- − The capybara's head has a slightly 'pasted-on' look with some artifacts around the neck.
- − The steering wheel appears to be emerging from the car door frame incorrectly.
- − Garbled text on the taxi roof sign.
Verdict: Seedream 4.5 captures the requested 'inside' perspective much better than Wan 2.7, creating a more immersion scene that feels directly from the car's interior. While Wan 2.7 has impressive anatomical detail on the paws, the composition and overall coherence of Seedream 4.5 make it the superior choice for this 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.”
Seedream 4.5
Wan 2.7
AI Judge Analysis
Seedream 4.5
- + Excellent photographic realism in the jack-o-lantern and mood lighting
- + Perfectly accurate text rendering for all requested details
- + Effective use of negative space for readability
- − The border elements like the barbed wire feel like separate overlays rather than an integrated part of a 'vintage poster'
- − Composition is slightly sparse compared to the 'vintage gothic' aesthetic requested
Wan 2.7
- + Outstanding interpretation of the 'vintage gothic' style with an intricate, hand-illustrated feel
- + Highly detailed border containing all requested elements (webs, thorns, bats) plus extra thematic additions
- + Text is beautifully integrated into the layout with a more appropriate gothic font
- − The jack-o-lantern and lighting are more illustrative/cartoonish than cinematic
- − The image includes extra text not requested in the prompt, such as 'Est. 1847' and 'Dress code'
Verdict: Seedream 4.5 produces a high-fidelity, modern take on the invitation with superior text clarity and realistic lighting. However, Wan 2.7 far better captures the 'vintage gothic' aesthetic with its complex, thematic border and stylized illustration, making it look much more like a real invitation despite some minor creative liberties with the text.
Next steps
Explore each model
Alibaba's Wan 2.7 image generation and editing model for text-to-image, reference-guided generation, and instruction-based image edits