Qwen Image Edit 2509 Alibaba Wan 2.7 Alibaba

Settled by community votes across 7 shared challenges, with an AI judge weighing in on each.

Qwen Image Edit 2509

7.8 arena score

#32 of 32 in Image Editing

Skill signature · Image Editing

Wan 2.7

20.5 arena score

#39 of 62 in Text-to-Image

Vote tally

Where the votes landed

Qwen Image Edit 2509

win rate

Ties

Wan 2.7

win rate

Shared challenges 7

Challenge by challenge

The strongest take from each model on every shared challenge, with the AI judge's read.

Man and Car in California

Editing

Edit instruction

“Make a photo of the man driving the car down the California coastline”

Source

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Excellent preservation of the car's details and reflections.
+ Accurately represents the man from the source image, including his unique hairstyle.
+ Strong sense of motion with high-quality blur on the road and wheels.

− The man's scale within the car seems slightly small.

Wan 2.7

+ Very realistic and scenic background composition.
+ Good integration of the car onto the road surface and lighting.

− Completely failed to use the man from the source image, substituting him with a generic figure.
− The face of the driver is distorted and poorly rendered.
− The car's front grille lacks the detail of the original source image.

Verdict: Qwen Image Edit 2509 is the clear winner as it successfully integrated the specific man from the source image into the car while maintaining his hairstyle and features. Wan 2.7 failed the primary task of placing the 'source man' in the vehicle, instead rendering a distorted, generic person and losing some fidelity in the car's details.

Pose & Character Mashup

Editing

Edit instruction

“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”

Source

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Successfully maintains the yellow background and red ottoman composition
+ Replicates the pose from Image 1 with high accuracy

− Completely failed to incorporate the character reference from Image 2
− Increased contrast to a point that looks artificial compared to the source

Wan 2.7

+ Perfectly preserves the source Image 1 as the final output
+ Retains the soft natural lighting and skin tones of the original pose image

− Complete failure to apply the requested edit
− Did not integrate any features (face, hair, clothing) from Image 2

Verdict: Both Qwen Image Edit 2509 and Wan 2.7 failed the primary task of cross-referencing Image 2's character with Image 1's pose. Both models simply returned variations of Image 1, ignoring the request to use the man from Image 2 as the subject. Wan 2.7 is slightly better only in that it did not degrade the image quality with the harsh contrast seen in Qwen Image Edit 2509.

Outfit Transfer Challenge

Editing

Edit 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

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Perfectly replicates the specific outfit from Image 2 including the coat, scarf, and jeans.
+ Successfully incorporates the sunglasses from Image 2.

− Failed the negative constraint by changing the person's face and build to look like the man in Image 2.
− Distorted the background and person's pose significantly.
− Low image resolution and poor blending at the edges.

Wan 2.7

+ Excellent preservation of the original person's face, hair, and unique skin features.
+ Maintains the background integrity and the person's original pose precisely.
+ High visual quality with realistic lighting and shadows on the clothing.

− Failed to use the exact outfit from Image 2, substituting it with a generic elaborate black and gold outfit instead.

Verdict: This is a trade-off between outfit accuracy and source preservation. Qwen Image Edit 2509 successfully transferred the correct clothes but failed completely at preserving the subject's identity, essentially pasting the second man's head and body into the scene. Wan 2.7 perfectly preserved the subject and background as requested, but failed to follow the instruction to use the 'exact' outfit, providing a creative alternative instead. Wan 2.7 is the likely winner for its superior professional quality and adherence to identity preservation.

Bald man challenge

Image Editing

Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Excellent preservation of the source background, jacket, and facial features.
+ Strong adherence to the 'thick head of hair' request.

− The hair flow and lighting on the scalp look slightly unnatural and stylized.
− The hair volume is arguably exaggerated for a 'natural' look.

Wan 2.7

+ Provides a very realistic, natural-looking haircut and hairline that blends seamlessly.
+ Perfectly preserves all details of the source image including clothing and background.
+ Realistic texture that matches the existing beard.

− The forehead area shows a very slight distortion where the new hairline meets the skin.

Verdict: Both models performed excellently at preserving the source image while adding hair. Wan 2.7 is the winner because the hair style and volume look much more grounded and realistic compared to the wavy, almost-caricatured volume produced by Qwen Image Edit 2509.

Over-the-top cartoon caricature

Editing

Edit instruction

“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”

Source

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Successfully incorporates all prompt elements: TV anchor desk, hockey rink background, and a dog.
+ Captures the 'caricature' style well with exaggerated facial features while maintaining a strong likeness to the source image.
+ Preserves the specific denim shirt from the original photo.

− Text in the speech bubble is nonsensical gibberish.
− The dog is very small and lacks detail compared to the rest of the image.

Wan 2.7

+ Includes more creative hockey/dog crossovers, such as a dog wearing a hockey helmet.
+ Text in speech bubbles is legible and thematic.
+ Excellent caricature style with expressive features and vibrant colors.

− Changes the subject's outfit from the denim shirt in the source to a professional blazer.
− The likeness to the original person is slightly less accurate than in Model A.

Verdict: Both models performed remarkably well, correctly capturing the requested caricature style and all thematic elements. Qwen Image Edit 2509 is slightly better at source preservation, keeping the subject's denim shirt and achieving a more precise facial likeness, though its text utility is poor. Wan 2.7 produced a more polished and humorous scene with readable text and fun details like the hockey-playing dog, making it the more engaging result despite the outfit change.

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

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Successfully captures the character art style of Ghibli films with simple lines and blush marks.
+ Excellent preservation of the source image composition and the specific expressions of the subjects.
+ Applies a clean, cell-shaded aesthetic consistent with modern digital anime.

− The halftone/reprographic texture overlay feels more like a manga print than a hand-painted Ghibli background.
− Significantly lightened the red dress to a pale pink, deviating from the source colors.

Wan 2.7

+ Captures the 'hand-painted textures' and 'soft pastel colors' requested in the prompt perfectly.
+ Beautiful watercolor-inspired background aesthetic that closely mimics Ghibli's scenery style.
+ Retains the original red color of the dress better than Model A.

− The character faces are a bit too realistic/detailed to truly feel like Ghibli character designs.
− Slightly less 'clean' than Model A in terms of line work.

Verdict: Both models do an excellent job of preserving the source image while applying the requested style. Qwen Image Edit 2509 excels at the character design aspect of Studio Ghibli, whereas Wan 2.7 captures the iconic painterly background textures and soft lighting much more effectively. Wan 2.7 is the winner for its superior 'hand-painted' feel and color preservation.

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.”

Qwen Image Edit 2509

Wan 2.7

AI Judge Analysis

Qwen Image Edit 2509

+ Excellent preservation of the original image identity and character features.
+ Clearly visible autumn leaves added throughout the frame.

− The wind effect on the hair is stylized and looks slightly unnatural/illustrative.
− The leaves are very sharp, lacking the motion blur needed to convey 'dynamic motion'.

Wan 2.7

+ Provides a more realistic wind effect on the hair with fine strands blowing naturally.
+ Adds motion blur to the flying leaves, creating a convincing sense of speed and energy.
+ Excellent preservation of the source image while integrating new elements.

− The leaves are slightly more sparse compared to the other model.

Verdict: Both models followed the instructions well, but Wan 2.7 is the clear winner due to the realism of its motion. While Qwen Image Edit 2509 added static, stylized hair and sharp leaves, Wan 2.7 successfully implemented motion blur on the leaves and a more natural wind effect on the hair, truly capturing the 'energetic and lively feel' requested.