GPT Image 2 OpenAI Qwen Image Edit 2509 Alibaba

Settled by community votes across 2 shared challenges, with an AI judge weighing in on each.

GPT Image 2

27.7 arena score

#4 of 62 in Text-to-Image

Skill signature · Text-to-Image

Qwen Image Edit 2509

7.8 arena score

#32 of 32 in Image Editing

Vote tally

Where the votes landed

GPT Image 2

win rate

Ties

Qwen Image Edit 2509

win rate

Shared challenges 2

Challenge by challenge

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

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

GPT Image 2

Qwen Image Edit 2509

AI Judge Analysis

GPT Image 2

+ Successfully replicates the exact pose and environment from Image 1.
+ Accurately migrates the character's face, sunglasses, and scarf from Image 2.
+ Matches the lighting and shadow of the original background perfectly.

− The character's left foot has an anatomical error with extra toes.
− Merging the scarf with the sweatshirt results in some minor textural inconsistencies.

Qwen Image Edit 2509

+ Maintains high color saturation and clarity.

− Completely failed the character reference task by simply returning a filtered version of Image 1.
− Failed to incorporate any elements (face, clothing, glasses) from Image 2.
− Produced a generic filter effect that reduces realistic detail.

Verdict: GPT Image 2 (Model A) followed the complex instructions excellently, successfully transplanting the specific character from Image 2 into the dynamic pose of Image 1 while maintaining consistent lighting and including key accessories like the scarf. Qwen Image Edit 2509 (Model B) failed the prompt entirely, providing a slightly edited version of the first source image without incorporating the second character at all.

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

GPT Image 2

Qwen Image Edit 2509

AI Judge Analysis

GPT Image 2

+ Perfect preservation of the original person's face, skin patterns, and expression.
+ Excellent integration of the clothing layers with realistic fabric folds and lighting.
+ Maintains the high resolution and texture of the original background.

− Does not include the sunglasses from Image 2.
− Crops the image, losing the full-length view requested to show shoes.

Qwen Image Edit 2509

+ Attempts to include more elements like the sunglasses and shoes.
+ Matches the blue saturation of the jeans well.

− Severely distorts the person's face, losing the original likeness and details.
− Poor composition with awkward body proportions and a 'floating' appearance.
− Low visual quality with significant artifacts and blurry textures.

Verdict: GPT Image 2 is the clear winner as it flawlessly preserves the identity and unique characteristics of the person from Image 1 while realistically applying the requested outfit. Qwen Image Edit 2509 fails significantly on source preservation, creating a distorted face that looks nothing like the original person and suffering from poor image coherence.