GPT Image 1.5 vs Z-Image Turbo
Head-to-head across 11 challenges
GPT Image 1.5
90.0%
win rate
Ties
10.0%
Z-Image Turbo
0.0%
win rate
Challenge Results
Geometric Composition
Text-to-Image“A glass cube on a wooden table. Inside the cube is a small blue sphere. On top of the cube sits a red book. A green plant is behind the cube, partially visible through the glass. Soft window light from the left.”
AI Judge Analysis
GPT Image 1.5
- + Excellent photographic detail on the book texture and wood grain.
- + Strong physical logic with the green plant clearly visible through the glass panels.
- + Accurate lighting and reflections on the glass and blue sphere.
- − The 'blue sphere' is quite large despite the prompt asking for a 'small' one.
Z-Image Turbo
- + Follows the 'small sphere' instruction better than Model A.
- + Clean composition with soft window light as requested.
- − Physics error where the plant in the background disappears behind the glass panes.
- − The glass cube handles reflections poorly, losing transparency in the back-right corner.
Verdict: GPT Image 1.5 is the superior image due to its consistent handling of transparency and reflections; the green plant is properly visible through the glass cube, whereas it magically vanishes in Z-Image Turbo. While Z-Image Turbo scaled the sphere more accurately to the prompt, its failure to maintain visual logic through the glass makes it less realistic.
Candid Street Photography
Text-to-Image“A candid street photo of an elderly Japanese man repairing a red bicycle in light rain, reflections on wet pavement, shallow depth of field, 50mm lens, natural skin texture, imperfect framing, motion blur from passing cars, cinematic but realistic, no stylization.”
AI Judge Analysis
GPT Image 1.5
- + Excellent adherence to the 'repairing' aspect of the prompt with tools and a crouched pose.
- + Superior atmospheric lighting and realistic reflections on wet pavement.
- + Perfect execution of the shallow depth of field and 'cinematic' look requested.
- − The bike anatomy is slightly jumbled near the rear wheel/derailleur area.
Z-Image Turbo
- + Clear representation of light rain and natural skin texture.
- + Good composition with a clear subject.
- − The man is standing/walking with the bike rather than repairing it.
- − The background cars lack the requested motion blur.
- − The lighting feels flat and less cinematic compared to the other image.
Verdict: GPT Image 1.5 followed the prompt much more closely, capturing the specific 'repairing' action and the cinematic atmosphere of a rainy street. While Z-Image Turbo produced a clean image, it missed the repair activity and the motion blur requirement, resulting in a more generic snapshot. GPT Image 1.5's use of light and reflections better captured the 'cinematic but realistic' tone requested.
Fantasy Warrior
Text-to-Image“Close portrait of a battle-worn paladin in ornate engraved plate armor, hair braided with small beads, faint scars and dirt on the skin, warm torchlight reflecting off metal, shallow depth of field, bokeh sparks, lifelike eyes, highly detailed texture on leather straps and cloth underlayer.”
AI Judge Analysis
GPT Image 1.5
- + Exceptional texture detail on the skin, scars, and engraved metal.
- + Stronger adherence to the 'close portrait' instruction with impactful framing.
- + Beautiful lighting and bokeh sparks that integrate naturally with the scene.
- − The hair beads are present but somewhat large and chunky compared to a delicate braid feel.
Z-Image Turbo
- + Successfully includes the specific bead details in the hair braids.
- + Captures the underlayer textures like chainmail and quilted fabric well.
- + Good use of the torch as a physical light source in the frame.
- − The composition is a medium shot rather than the requested 'close portrait'.
- − The skin texture and facial features lack the hyper-realistic clarity seen in Model A.
- − Lighting on the face is a bit flat despite the presence of the torch.
Verdict: GPT Image 1.5 produced a far more compelling and high-quality image that perfectly matches the 'close portrait' and 'lifelike' requirements of the prompt. While Z-Image Turbo followed the instructions for hair beads and leather/cloth layers well, the overall image quality and resolution in GPT Image 1.5 are superior, offering much more intricate detail in the skin and metal textures.
Modern Clean Menu
Text-to-Image“Modern minimalist restaurant menu design, white background with colorful food photos in grid, sections for appetizers/pizza/mains, bold sans-serif fonts, vibrant accents, clean professional layout for casual dining.”
AI Judge Analysis
GPT Image 1.5
- + Perfectly legible text with correct spelling and descriptions.
- + Excellent alignment between text sections and corresponding food imagery.
- + Very high-quality, realistic food photography that looks professional.
- − The layout is a bit standard, leaning more towards a flyer than a modern square grid design.
Z-Image Turbo
- + Follows the 'grid' request more literally with a structured tile layout.
- + Bold use of color blocks and sans-serif fonts matches the 'vibrant accents' prompt.
- − Text is largely nonsensical gibberish (e.g., 'PIZZA MANS', 'SE TIIION').
- − Food photos are repetitive and lower in visual quality compared to Model A.
- − Poor hierarchy and spacing in the text columns.
Verdict: GPT Image 1.5 is the clear winner because it produces a functional, professional menu with perfectly rendered text and high-quality food photography. Z-Image Turbo followed the 'grid' instruction well, but failed significantly on text legibility and overall image coherence.
Magic Burger Explosion: Fiery Photorealism Challenge
Text-to-Image“Ad for 'Magic Burger'. Dynamic, exploded burger with all components (bun, patty, cheese, lettuce, tomato, sauce) suspended in mid-air. Emphasize photorealistic detail and a sense of motion. Dark, fiery background with glowing embers. Integrate text: 'MAGIC BURGER' as a prominent title, 'LIMITED TIME ONLY' as a secondary message, and '€6.99' in a starburst, all rendered with a fiery, glowing effect.”
AI Judge Analysis
GPT Image 1.5
- + Excellent 'exploded' effect with all components physically separated as requested.
- + Highly detailed and photorealistic food textures, especially the lettuce and sauce.
- + Superior integration of the fiery glowing text effects across all three required text elements.
- − The background is slightly busy, which may distract from the product details.
Z-Image Turbo
- + Clean composition with a clear focus on the product and price tag.
- + Accurate text spelling for all requested phrases.
- + Professional lighting on the burger bun and patties.
- − Failed to deliver an 'exploded' burger, showing a fully assembled sandwich instead.
- − Repetitive text 'MAGIC BURGER BURGER' creates an error in the main title.
- − The background feels static rather than dynamic and motion-filled.
Verdict: GPT Image 1.5 followed the prompt much more accurately, successfully creating the difficult 'exploded' burger effect while maintaining incredible photorealistic detail. Z-Image Turbo ignored the core request for an exploded view and included a redundant word in the title, resulting in a less dynamic advertisement.
Chalkboard Menu
Text-to-Image“Handwritten-style chalkboard menu in a cozy café, all text rendered in the exact same realistic chalk handwriting style with natural variations in letter size, slight slant, and chalk texture — no printed or digital fonts anywhere on the board. Title at the top in elegant cursive chalk handwriting: ‘TODAY’S SPECIALS – APRIL 30, 2026’. Below it, three menu items also in the same handwritten chalk style: ‘Truffle Mushroom Risotto – $24’, ‘Grilled Octopus with Lemon & Herbs – $28’, ‘Brown Butter Chocolate Chip Cookies – $9’. At the very bottom, smaller text in the identical handwritten chalk style (slightly smaller but still clearly legible with the same handwriting characteristics): ‘All items made fresh daily • Ask about our gluten-free options’. Warm ambient café lighting, visible chalk dust and smudges, realistic handwriting imperfections, no clean printed text anywhere.”
AI Judge Analysis
GPT Image 1.5
- + Excellent chalk texture with realistic smudging and dusting on the board.
- + Flawless spelling and adherence to the handwriting style requested.
- + Superior cursive rendering for the title while maintaining a consistent hand for the list.
- − The composition leaves a large amount of empty space at the bottom.
Z-Image Turbo
- + Good use of the available vertical space on the chalkboard.
- + Clear and legible text rendering.
- − Includes a spelling error with 'Mustroom' instead of 'Mushroom'.
- − The chalk texture looks more like a digital white brush than actual chalk.
- − Failed the 'elegant cursive' requirement for the title, using a standard print style instead.
Verdict: GPT Image 1.5 is the clear winner as it perfectly followed all stylistic and content instructions, providing a highly realistic chalk texture and accurate handwriting. Z-Image Turbo failed to provide the cursive title requested, produced a spelling error, and had a less convincing texture that appeared more like a digital font.
The Capybara Taxi Driver
Text-to-Image“Photorealistic scene inside a yellow New York taxi at night. A capybara is driving, wearing a yellow taxi driver cap and a dark jacket. It has a calm, professional expression and both front paws on the steering wheel. In the back seat sits a human businesswoman in a coat, looking at her phone with a completely normal, bored expression (as if this is just another normal ride). Through the windows you can see the streets of Manhattan at night with blurred lights. Realistic taxi interior, photorealistic, detailed fur and fabric, 35mm lens, night lighting with reflections, shallow depth of field.”
AI Judge Analysis
GPT Image 1.5
- + Excellent photorealism with gritty, cinematic lighting that feels like a real New York taxi at night.
- + The text on the taxi cap is clear and accurate.
- + Superior integration of the capybara's anatomy with the steering wheel and seatbelt.
- − The passenger in the back is slightly out of focus, though this helps with the sense of depth.
Z-Image Turbo
- + Good adherence to all major prompt elements like the capybara driver and bored passenger.
- + Clean, high-resolution image with vibrant colors.
- − The capybara's paws are not actually on the steering wheel, appearing to float or grasp air behind it.
- − The lighting and textures feel slightly more like a digital render than a photograph.
- − The passenger holds the phone in a slightly unnatural way.
Verdict: GPT Image 1.5 is the clear winner due to its superior photorealistic quality and attention to physical interaction, such as the paws realistically gripping the steering wheel. Z-Image Turbo captures the concept well but fails on the specific detail of the paws on the wheel and has a less convincing cinematic atmosphere.
The Halloween Invitation
Text-to-Image“Vintage gothic Halloween party invitation. Dark parchment poster, spooky border with webs and thorns, central glowing jack-o-lantern, bats, twisted trees, moody night sky. Add elegant gothic title text saying "Halloween Party Invitation", a small scroll banner saying "You are invited to a night of frights", and event details at the bottom: Date: 30.10.2026 Time: 7pm Location: The Arches, NYC Spooky but polished, cinematic lighting, square format.”
AI Judge Analysis
GPT Image 1.5
- + Excellent typography with zero spelling errors.
- + Perfect aesthetic for dark parchment with a vintage, atmospheric feel.
- + Thoroughly follows all layout instructions including the banner and event details.
- − The thorns in the border look a bit more like dried branches than distinct thorns.
- − The text on the scroll banner is a bit large compared to the banner size.
Z-Image Turbo
- + Strong parchment-on-background layered effect.
- + Good incorporation of the twisted trees and moody sky elements.
- + High contrast and clarity in the center jack-o-lantern.
- − Typos in the location text (e.g., 'The Archves') and title alignment.
- − Failed to include the specific text 'You are invited to a night of frights' on the scroll banner, placing it at the very top instead.
- − The overall composition feels a bit more like digital clip-art than a cohesive vintage poster.
Verdict: GPT Image 1.5 is the clear winner as it followed all prompt instructions perfectly, particularly regarding the text content. While Z-Image Turbo created a visually interesting layered effect, it suffered from spelling errors and missed the instruction to place specific text on the banner. GPT Image 1.5 also better captured the 'cinematic' and 'vintage gothic' mood requested.
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
GPT Image 1.5
- + Expertly follows the prompt providing a full, thick head of hair with natural-looking curls.
- + Maintains excellent source preservation of the facial features, clothes, and overall lighting.
- + The integrated hair texture matches the existing beard density and style perfectly.
- − Slight adjustment to the top frame of the glasses near the nose bridge compared to the original.
Z-Image Turbo
- + Maintains the overall composition and color palette of the original photo.
- − Completely failed the primary edit request, leaving the subject bald with only a slight stubble shadow.
- − Lost the subject's glasses entirely, which was not requested.
- − Significantly altered the facial features and the background landscape.
Verdict: GPT Image 1.5 successfully executed the edit by adding a realistic, thick head of hair that blends seamlessly with the original subject's features and lighting. Z-Image Turbo failed the prompt entirely, failing to add hair while also mistakenly removing the subject's glasses and altering the background.
Adorable Baby Animals in Sunny Meadow
Text-to-Image“Hyper-photorealistic scene of fluffy baby animals—a golden retriever puppy, tabby kitten, baby bunny, and red fox kit—with big expressive eyes and ultra-detailed soft fur, playfully chasing butterflies and tumbling together in a lush wildflower meadow, warm golden sunrise light with god rays and dew sparkles, joyful wholesome vibe, 8K masterpiece.”
AI Judge Analysis
GPT Image 1.5
- + Excellent adherence to lighting requests with clear god rays and dew sparkles.
- + Highly expressive and dynamic poses that match the 'tumbling' and 'chasing' aspects of the prompt.
- + Superior texture rendering in the fur and floral elements.
- − The kitten's anatomy, specifically its paws and belly, looks slightly distorted.
- − The background is very busy, which slightly distracts from the subject.
Z-Image Turbo
- + Clear and distinct representation of all four requested animals.
- + Clean composition with a soft, pleasant bokeh effect.
- + Good anatomical consistency for all animals.
- − Lighting is relatively flat and lacks the requested 'god rays' and 'dew sparkles'.
- − The animals look more like they are posing than 'tumbling' or 'playfully chasing'.
- − The butterflies appear somewhat static and floaty.
Verdict: GPT Image 1.5 captured the atmosphere of the prompt much more effectively, providing the requested god rays, dew sparkles, and a genuine sense of dynamic 'tumbling' movement. While Z-Image Turbo produced a cleaner, more anatomically stable image, it failed to incorporate the specific environmental effects requested and felt more like a posed studio shot than a masterpiece scene.
Vintage Cafe Logo
Text-to-Image“Vintage minimalist restaurant logo for "Caffè Florian", retro cloche dome with steam and "Est. 1720" banner, classic typography, warm brown and cream tones, subtle texture on light background, vector emblem style.”
AI Judge Analysis
GPT Image 1.5
- + Excellent typography style that feels premium and vintage
- + High quality shading and grain texture on the cloche dome
- + Perfect spelling of names and dates
- − Ignored the request for a light background, providing a black one instead
- − Slightly less 'minimalist' than requested due to heavy shading
Z-Image Turbo
- + Followed the light background and subtle texture instructions perfectly
- + True minimalist vector aesthetic
- + Clean, balanced layout
- − The 'steam' element is very small and lacks artistic impact
- − Typography is more modern/basic compared to the requested 'classic' feel
Verdict: Z-Image Turbo adhered more closely to the full prompt by providing the requested light background and minimalist vector style, whereas GPT Image 1.5 failed on the background color despite producing a more visually impressive and detailed vintage illustration. If the user requires a usable logo on a light surface as requested, Z-Image Turbo is the functional winner, though GPT Image 1.5 has superior artistic execution for a vintage brand.
GPT Image 1.5
OpenAI's state-of-the-art image generation model with better instruction following and adherence to prompts
Z-Image Turbo
Tongyi-MAI's 6-billion parameter distilled text-to-image model optimized for speed, achieving high-quality generation in 8 steps or fewer with support for bilingual text rendering