GPT Image 1.5 OpenAI P-Image Edit PrunaAI

Settled by community votes across 1 shared challenge, with an AI judge weighing in on each.

GPT Image 1.5

26.5 arena score

#7 of 44 in Text-to-Image

Top 3 in Image Editing

Skill signature

Not enough comparable category data

The chart appears once both models have ratings across at least three shared arena categories.

P-Image Edit

0.2 arena score

#24 of 23 in Image Editing

Vote tally

Where the votes landed

GPT Image 1.5

win rate

Ties

P-Image Edit

win rate

Shared challenges 1

Challenge by challenge

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

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 1.5

P-Image Edit

AI Judge Analysis

GPT Image 1.5

+ Excellent fabric texture and realistic clothing weight
+ Very high fidelity to the original scarf and coat patterns
+ Natural lighting and seamless integration with the wood pillar

− Cropped the person's head, failing to show the exact face and hair as requested
− Loss of the original image's high-contrast beach lighting

P-Image Edit

+ Successfully kept the full person including head and unique vitiligo features
+ Preserved the background and framing of Image 1 much better than Model A
+ Incorporated accessories like the watch and jewelry

− Lower visual quality with visible artifacts around the skin and hair
− The shoes and gold chain feel like digital stickers rather than integrated objects
− Proportions of the legs appear slightly warped compared to the original torso

Verdict: GPT Image 1.5 achieved a much more realistic clothing transfer with superior textures and lighting, but it failed the fundamental requirement of keeping the person's face and hair by cropping them out. P-Image Edit followed the composition instructions much better, keeping the subject's identifying features intact, though the technical quality of the 'photoshop' is significantly lower and contains noticeable artifacts. P-Image Edit is the winner for actually completing the task, whereas Model A produced a high-quality image that ignored the subject preservation requirement.