P-Image Edit PrunaAI vs Qwen Image Edit 2511 Alibaba

P-Image Edit

• + Excellent preservation of the original car model and orientation.
• + Convincing motion blur on the road and wheels creates a sense of driving.
• + Smooth integration of the subject into the driver's seat.

• − The man is very small and distant in the frame.
• − The character's clothing from the source image is lost.

Qwen Image Edit 2511

• + High fidelity to the man's facial features and distinctive hairstyle.
• + Excellent preservation of the source clothing (plaid coat and scarf).
• + Cinematic composition with clear view of both the subject and the coastline.

• − The steering wheel and dashboard are modified from the original Rolls Royce.
• − The car appears to be parked or stationary rather than driving due to the lack of motion blur.

P-Image Edit

• + Successfully merges the character from Image 2 with the pose from Image 1.
• + Incorporates the scarf and sunglasses from the character reference into the final pose.
• + Maintains the environment and lighting of the pose source image.

• − Anatomy is nonsensical with the leg extending directly out of the head/shoulder area.
• − The long hair from Image 1 was kept instead of using the short hair from the character in Image 2.

Qwen Image Edit 2511

• + High resolution and clear image quality.

• − Failed the main task by simply layering the two images on top of each other.
• − The character from Image 2 is not performing the pose; they are just standing in the background.
• − Source preservation is poor as the background and character are just composited together blindly.

P-Image Edit

• + Successfully preserved the unique vitiligo patterns on the person's face.
• + Accurately captured the specific plaid scarf and coat style from Image 2.
• + Followed the instruction to keep the base person's identity perfectly.

• − The shoes are an inaccurate addition not found in either source image.
• − Included a large gold chain that was not in Image 2.
• − The lighting on the person's skin is slightly washed out compared to the original.

Qwen Image Edit 2511

• + Excellent integration of the clothing onto the model with realistic shadows.
• + Captured the pose and lean of the original image well.

• − Completely failed to preserve the person's face, replacing them with the man from Image 2.
• − Ignored the instruction to keep the person's exact hair and vitiligo patterns.
• − Misinterpreted 'keep the person's exact face... unchanged' entirely.

P-Image Edit

• + Successfully added high-density, curly hair as requested.
• + Preserved the background and clothing perfectly.
• + The lighting on the curls matches the environment reasonably well.

• − The hairline is very abrupt and lacks natural transition/gradience.
• − The curls have a slightly repetitive, digital appearance toward the edges.
• − Changed the shape of the upper forehead slightly too much.

Qwen Image Edit 2511

• + Created a much more realistic, natural-looking hairline with subtle baby hairs.
• + The hair texture and wave patterns are highly realistic and varied.
• + Excellent preservation of the original facial features and lighting integration.

• − The hair volume is slightly lower than Model A, though still fits the prompt.

P-Image Edit

• + Successfully incorporates all elements: TV anchor, dog, and hockey players.
• + Maintains the subject's denim shirt from the original photo as requested for source preservation.
• + Cleverly sets the TV desk in front of a hockey rink backdrop.

• − The dog has a strange red heart-shaped structure growing out of its back.
• − The hand holding the microphone has some anatomical distortion.

Qwen Image Edit 2511

• + Captures a strong, humorous caricature facial expression while maintaining likeness.
• + The dog is rendered very cleanly and integrates well into the scene.
• + The composition is clean and follows the 'TV anchor' prompt closely.

• − Completely misses the 'hockey' requirement of the prompt.
• − The text in the speech bubble is nonsensical gibberish.
• − A strange candle on the desk feels out of place with the modern studio setting.

P-Image Edit

• + Successfully captures the requested dreamlike background with fluffy clouds and gardens.
• + Maintains the distinct poses and layout of the original meme perfectly.
• + The color palette is vibrant yet soft, fitting the requested nostalgic mood.

• − The facial expressions are slightly softened, losing some of the sharp intensity of the woman on the right.

Qwen Image Edit 2511

• + Preserved the background architecture of the original image with high accuracy.
• + Captures the facial expressions, especially the intense glare of the woman on the right, more accurately.
• + Excellent hand-painted textures on the clothing and skin.

• − The background is less 'dreamy' and Ghibli-esque than Model A, retaining more of the realistic urban street.
• − Includes some small white artifacts/sparkles scattered across the image that weren't requested.

P-Image Edit

• + Excellent preservation of the woman's face and body from the source image
• + High identity consistency for the dog
• + Accurately places many small flying leaves throughout the scene

• − The hair edit appears slightly more stiff/less flowing compared to Model B

Qwen Image Edit 2511

• + Very natural and energetic hair motion that looks well-integrated
• + Leaves are varied in size and color, adding to the depth of the scene
• + Exceptional preservation of the source image details and person's likeness

• − Minor distortion in the dog's left eye compared to the source