GPT Image 1.5 OpenAI HiDream E1 HiDream AI

Settled by community votes across 4 shared challenges, with an AI judge weighing in on each.

GPT Image 1.5

27.1 arena score

#7 of 62 in Text-to-Image

Top 3 in Image Editing

Skill signature · Text-to-Image

HiDream E1

0.7 arena score

#33 of 32 in Image Editing

Vote tally

Where the votes landed

GPT Image 1.5

win rate

Ties

HiDream E1

win rate

Shared challenges 4

Challenge by challenge

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

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.”

GPT Image 1.5

HiDream E1

AI Judge Analysis

GPT Image 1.5

+ Perfect preservation of the original facial features and body.
+ Superior integration of the new hair with the existing beard and skin.
+ Highly realistic texture and natural-looking hairline.

− None identified.

HiDream E1

+ Followed the instruction to add a full head of hair.

− Failed to preserve the background, changing the desert landscape significantly.
− Hair texture looks synthetic and lacks realistic integration with the scalp.
− Skin texture and lighting were altered, losing the original photographic quality.

Verdict: GPT Image 1.5 performed an nearly flawless edit, seamlessly integrating a realistic head of hair while preserving every detail of the original photo. In contrast, HiDream E1 fundamentally changed the entire image, including the background and skin texture, while the added hair looks like a digital overlay rather than a natural part of the person.

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

GPT Image 1.5

HiDream E1

AI Judge Analysis

GPT Image 1.5

+ Excellent adherence to all prompt elements including dogs, hockey, and a news anchor desk.
+ High-quality caricature style that successfully preserves facial features from the source image.
+ Creative and humorous integration of the dog wearing a hockey helmet.

− The hand holding the microphone has some anatomical distortion/extra fingers.

HiDream E1

+ Strong caricature exaggeration with expressive eyes and smile.
+ Clear integration of a hockey rink background.

− Significantly fails to preserve the source subject's likeness, changing hair color and facial structure.
− The rendering of the dog and hands is quite messy with blurring and anatomical errors.
− Includes small, blurry figures in the background that detract from the focus.

Verdict: GPT Image 1.5 is the clear winner as it successfully maintains the subject's identity from the source image while creatively incorporating the news desk, dogs, and hockey elements into a cohesive caricature. HiDream E1 fails to preserve the subject's likeness and suffers from significant anatomical and technical artifacts, particularly in the rendering of the hands and the dog.

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

GPT Image 1.5

HiDream E1

AI Judge Analysis

GPT Image 1.5

+ Excellent adherence to Ghibli's hand-painted watercolor textures.
+ Preserves the exact poses, clothing patterns, and expressions from the source image.
+ Rich, dreamy lighting and a nostalgic color palette.

− The woman in the foreground is slightly more blurred than in the original image.

HiDream E1

+ Successfully adopts a clean anime-style line art.
+ Maintains the basic composition of the original meme.

− Loses significant character details, such as the man's plaid shirt and the girlfriend's angry expression.
− The art style feels more like a generic modern webcomic than the requested lush Ghibli aesthetic.
− Background detail is overly simplified and flat.

Verdict: GPT Image 1.5 is the clear winner as it perfectly captures the specific Studio Ghibli 'hand-painted' look with soft textures and dreamy lighting while maintaining almost perfect fidelity to the original image's character details. HiDream E1 produces a much simpler, flatter illustration that loses the iconic plaid patterns and the emotional nuances of the original expressions.

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.”

GPT Image 1.5

HiDream E1

AI Judge Analysis

GPT Image 1.5

+ Excellent preservation of the original subjects' features and faces.
+ Realistic hair movement that fans out naturally.
+ High density of leaves creates a strong sense of wind.

− The leaf colors (yellow/brown) look slightly autumnal despite the lush green trees.
− Some leaves appear slightly blurry or like artifacts due to the quantity.

HiDream E1

+ Clean hair motion that flows in a consistent direction.
+ Leaves have high contrast and are easily visible.

− Significantly altered the woman's facial features and bone structure.
− The dog appearance changed noticeably, looking more stylized and less like the source.
− Leaf placement feels artificial and disconnected from the environment.

Verdict: GPT Image 1.5 is the clear winner as it successfully applied all requested motion effects while maintaining near-perfect source preservation of the woman and her dog. HiDream E1 failed the primary objective of an edit task by significantly altering the identities of the subjects, making them look like different characters than the ones in the source image.