PrunaAI's sub-1-second multi-image editing model supporting up to 5 reference images with state-of-the-art quality
Settled by community votes across 7 shared challenges, with an AI judge weighing in on each.
P-Image Edit
#29 of 32 in Image Editing
Wan 2.7 Pro
#38 of 62 in Text-to-Image
Where the votes landed
P-Image Edit
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.
Man and Car in California
Editing“Make a photo of the man driving the car down the California coastline”
AI Judge Analysis
P-Image Edit
- + Excellent preservation of the specific man from the source image, including his hairstyle.
- + Maintains the exact car model and design with high fidelity.
- + Effective motion blur on the wheels and road suggests realistic driving.
- − The man is positioned a bit high and small within the cockpit.
- − Slightly less atmospheric lighting compared to the alternative.
Wan 2.7 Pro
- + Beautiful evening lighting and coastal scenery.
- + Solid preservation of the car's geometry and reflections.
- − Completely failed to use the man provided in the source image, replacing him with a different person.
- − The car appears to be parked or static despite the prompt asking for him to be driving.
Verdict: P-Image Edit successfully combined both source images, correctly placing the specific man with the dreadlock hairstyle behind the wheel of the white Rolls Royce. Wan 2.7 Pro failed the core identity preservation task by substituting the man for a generic person, and the car looks stationary in its output.
Pose & Character Mashup
Editing“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.”
AI Judge Analysis
P-Image Edit
- + Successfully integrated character elements like the face, sunglasses, scarf, and black clothing.
- + Matches the dynamic pose from Image 1 effectively.
- + Maintains the yellow background and red platform from the source environment.
- − Anatomical failure with the raised leg appearing to come out of the side of the torso.
- − Bizarre inclusion of long flowing hair that does not match the short-haired character in Image 2.
- − Lighting on the face is inconsistent with the rest of the body.
Wan 2.7 Pro
- + Perfectly preserves the composition and visual quality of Image 1.
- − Completely failed the edit instruction by returning the source Image 1 with no character changes.
- − Ignored all details from the character reference in Image 2.
Verdict: P-Image Edit attempted the complex task of merging the two images, successfully pulling in the facial features, accessories, and clothing style of Image 2 into the pose of Image 1, though it suffered from significant anatomical distortions and added incorrect hair. Wan 2.7 Pro failed the task entirely by simply reproducing the first source image without any modifications. Therefore, P-Image Edit is the winner for actually attempting the edit, despite its technical flaws.
Outfit Transfer Challenge
Editing“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.”
AI Judge Analysis
P-Image Edit
- + Perfectly captured the exact outfit from Image 2 including the navy peacoat, plaid scarf, and blue jeans
- + Maintained the subject's unique skin patterns and facial features reasonably well
- + Correctly interpreted the pose and adjusted the clothing layers to fit the vertical full-body composition
- − Added bright yellow laces to the shoes which were not in either source image
- − The resolution is lower than the original source image, resulting in a slightly soft appearance
Wan 2.7 Pro
- + Maintained a higher image resolution and sharper details in the background and wood textures
- + Successfully preserved the skin patterns on the hands and chest
- − Completely failed to use the specified outfit from Image 2, choosing a gold/black patterned blazer instead
- − Failed to include the scarf, sunglasses, or specific style of the reference clothing
Verdict: P-Image Edit successfully followed the complex instruction to transfer a specific outfit from one image to another, maintaining the subject's identity and the core pieces of clothing (peacoat, scarf, jeans). While Wan 2.7 Pro produced a high-quality image with better clarity, it completely ignored the specific visual reference for the outfit, providing a random stylized suit instead.
Bald man challenge
Image Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
P-Image Edit
- + Excellent preservation of the background and original clothing details
- + Successfully added a massive volume of hair as requested
- − Hair looks somewhat artificial and 'pasted on' with a repetitive curl pattern
- − The hairline and edges of the hair look slightly inconsistent with the lighting
Wan 2.7 Pro
- + Extremely realistic hair texture and integrated lighting
- + The hairline looks incredibly natural and matches the person's age and facial structure
- − Slightly less volume than Model A, though still 'full'
- − Very minor blurring near the crown of the head
Verdict: Wan 2.7 Pro is the clear winner for its superior integration of the new hair, which looks like it belongs to the subject in terms of lighting and texture. While P-Image Edit provides more volume, the result looks like a digital overlay, whereas Wan 2.7 Pro creates a convincing and realistic transformation.
Over-the-top cartoon caricature
Editing“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.”
AI Judge Analysis
P-Image Edit
- + Excellent caricature translation that maintains the recognizable likeness of the source image.
- + Rich, detailed composition that blends a TV studio with a hockey rink seamlessly.
- + High-quality cartoon rendering with vibrant colors and clear, relevant background elements.
- − The 'dog' element is a bit small and strange, looking like a toy with an accessory rather than a real pet.
- − Text on screens in the background is garbled.
Wan 2.7 Pro
- + Strong 'caricature' energy with exaggerated facial expressions.
- + Uses clever props like a microphone/hockey stick hybrid and dogs in jerseys.
- + Clear, legible text in speech bubbles.
- − The facial likeness to the source subject is poorly preserved compared to Model A.
- − The background is relatively generic and less detailed than its competitor.
Verdict: P-Image Edit is the clear winner because it successfully transforms the subject into a caricature while maintaining a strong visual likeness to the source photo. While Wan 2.7 Pro has clever ideas for the props, it fails to preserve the identity of the person, creating a generic character instead of a caricature of the specific user.
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
P-Image Edit
- + Successfully captures the Studio Ghibli 'dreamy' background aesthetic with fluffy clouds and blooming trees.
- + Preserves the composition while bringing the foreground figure into focus as requested by the style.
- + Uses clean line work and soft pastel colors characteristic of modern anime.
- − Replaces the urban street background with a generic park, losing the specific context of the original photo.
- − Character features look a bit too generic and less like the original people compared to Model B.
Wan 2.7 Pro
- + Executes a beautiful hand-painted watercolor texture that feels authentic to Ghibli concept art.
- + Maintains the architectural background of the source image while stylizing it perfectly.
- + Preserves the likenesses and expressions of the subjects much more accurately than Model A.
- − The background is a bit washed out, losing some of the 'dreamy' vibrancy requested in the prompt.
Verdict: Wan 2.7 Pro is the clear winner for its superior ability to preserve the identity of the people and the environmental context while applying a sophisticated watercolor aesthetic. P-Image Edit creates a more generic 'anime' look and replaces the background entirely, whereas Wan 2.7 Pro successfully transforms the existing scene into a high-quality illustration.
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
P-Image Edit
- + Excellent preservation of the original image's lighting and background details.
- + Follows the instruction to add blowing hair and falling leaves accurately.
- + The hair motion feels symmetrical and dramatic.
- − The leaves appear a bit flat and like digital overlays rather than integrated into the scene.
- − The hair has some unnatural thin strands extending very far outwards.
Wan 2.7 Pro
- + Natural integration of leaves with motion blur and varying sizes, creating a better sense of depth.
- + Effective hair motion that feels slightly more grounded and realistic than Model A.
- + Successfully adds a glowing, energetic warmth to the lighting on the left side.
- − Slightly alters the facial features and hair color of the subject compared to the original.
- − Introduces a few artifacts where leaves overlap with the background trees.
Verdict: Both models followed the instructions well, applying wind effects to the hair and adding falling leaves. P-Image Edit was superior at preserving the exact likeness and background of the source image, while Wan 2.7 Pro created a more believable 'energetic' atmosphere through better leaf rendering and lighting adjustments, despite a slight loss in subject fidelity.
Explore each model
Alibaba's Wan 2.7 Pro image generation and editing model with higher-quality outputs and support for 4K image generation