GPT Image 1.5 vs Imagen 4.0 Ultra Generate 001
Head-to-head across 6 challenges
GPT Image 1.5
75.0%
win rate
Ties
8.3%
Imagen 4.0 Ultra Generate 001
16.7%
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 adherence to the 'small' scale of the sphere relative to the cube.
- + Very realistic rendering of thick glass edges and reflections.
- + Natural lighting and high-quality textures on the table and book cover.
- − The green plant in the background is quite cluttered compared to the clean aesthetic of the foreground.
Imagen 4.0 Ultra Generate 001
- + Beautiful bokeh effect and professional photographic composition.
- + Impressive text rendering on the book spine ('ANCIENT TALES').
- + Sophisticated handling of light and shadow across the wooden table.
- − The blue sphere appears to be floating rather than sitting inside the cube.
- − The cube looks more like a solid glass block than a hollow container.
Verdict: GPT Image 1.5 followed the prompt instructions more accurately, particularly regarding the placement and scale of the sphere inside the glass cube. While Imagen 4.0 Ultra Generate 001 produced a more artistic and visually stunning image with impressive text rendering, the central subject looks like a solid block with a floating sphere, failing the physics of the requested scene.
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
- + Captures motion blur in the passing car perfectly as requested.
- + The 50mm shallow depth of field is well-executed with professional-looking bokeh.
- + Excellent environmental storytelling with the toolkit and damp clothing textures.
- − The bicycle mechanics are slightly physics-defying with the rear rack and chain alignment.
- − The red reflection on the pavement is a bit oversaturated compared to the ambient light.
Imagen 4.0 Ultra Generate 001
- + Incredible skin texture and facial detail, looking very realistic and non-stylized.
- + Highly detailed rain droplets on the man's jacket.
- + Excellent composition using the brick wall as a leading line.
- − Failed to include motion blur on the passing car, which appears static.
- − The bicycle frame and pedals have some structural inconsistencies where they meet the chainring.
Verdict: GPT Image 1.5 adhered better to the technical requirements of the prompt, specifically the motion blur and the overall 'candid street photo' feel. While Imagen 4.0 Ultra provided superior skin textures and sharp details on the clothing, it missed the motion blur instruction and felt slightly more posed.
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
- + Perfect text rendering with zero spelling errors
- + Excellent logical organization of categories and pricing
- + High-quality, appetizing food photography that matches the text descriptions
- − Simple layout is highly functional but follows a very standard template style
Imagen 4.0 Ultra Generate 001
- + Stronger adherence to the 'grid' request in the prompt
- + Clean, modern minimalist aesthetic with plenty of white space
- − Garbled, nonsensical text in titles and descriptions
- − Confusing grouping where pizza items appear under 'Appetizers'
- − Inconsistent image sizes and alignment within the grid
Verdict: GPT Image 1.5 is the clear winner as it produces a fully functional, professional menu with perfect text and logical content mapping. While Imagen 4.0 Ultra attempted a more sophisticated grid layout, it failed significantly on text legibility and categorized the food items incorrectly.
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 texture work on the fur and paws, feeling very tactile and soft.
- + Beautiful lighting with visible god rays and realistic dew sparkles on the flowers.
- + Captures the 'tumbling together' part of the prompt with a natural, intertwined composition.
- − The fox kit's anatomy looks a bit distorted, especially the mouth and eye alignment.
Imagen 4.0 Ultra Generate 001
- + Very clear and vibrant colors with distinct butterfly designs.
- + Accurate representation of all four requested animals with clear silhouettes.
- + Good usage of dew drops on the grass and strong lighting rays.
- − Has a notably 'digital illustration' look rather than the requested 'hyper-photorealistic' style.
- − The animals feel more like they are floating or standing in place rather than 'tumbling together'.
Verdict: GPT Image 1.5 followed the stylistic requirements much better, delivering a truly photorealistic scene with soft, detailed textures and a natural, chaotic sense of play. Imagen 4.0 Ultra Generate 001 produced a very clean and cute image, but it leans heavily into a CGI/illustrative aesthetic rather than the realism requested by the prompt.
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 with correct accent on 'Caffè'
- + Strong vector illustration style with attractive shading
- + Clear and legible ribbon banner
- − Ignored the request for a light background with subtle texture
- − Steam effect is a bit chunky compared to the overall logo style
Imagen 4.0 Ultra Generate 001
- + Perfectly follows the light background with subtle texture requirement
- + Captures a more minimalist, clean aesthetic
- + Accurate text rendering and placement
- − The 'E' in 'Caffè' has a slightly awkward tail
- − Steam lines are very thin and faint compared to other elements
Verdict: Imagen 4.0 Ultra Generate 001 is the winner because it followed all prompt instructions, including the specific request for a textured light background. While GPT Image 1.5 produced a high-quality graphic with superior illustration details, it completely ignored the background color requirement, resulting in a black background.
Apollo 11: Journey to Tranquility
Text-to-Image“Create a clean, modern vector infographic poster about the Apollo 11 mission. NASA-inspired palette (navy, white, muted red, light gray). Flat-vector style, crisp lines, consistent iconography, subtle gradients only. Steps (stop at landing): 1. Launch (Saturn Vicon) 2. Earth Orbit (Earth + orbit ring icon) 3. Translunar (trajectory arc icon) 4. Lunar Orbit (Moon + orbit ring icon) 5. Descent (lunar module descending icon) 6. Landing (lunar module on the surface icon) Small supporting elements (minimal text): • Crew strip: three silhouette icons with only last names: Armstrong, Aldrin, Collins. • Landing site marker: Moon pin labeled "Tranquility" only. Layout constraints: generous margins, large readable labels, clean background with subtle stars. Vector-only, print-poster look, high resolution.”
AI Judge Analysis
GPT Image 1.5
- + Strictly followed the requested 6-step chronological sequence with accurate icons.
- + Excellent text rendering for main labels and supporting details like astronaut names.
- + Perfect adherence to the flat-vector style and specified NASA color palette.
- − Includes a redundant Earth icon in the first 'Launch' step which wasn't specifically requested.
- − The Saturn V rocket is slightly cut off at the top of the frame.
Imagen 4.0 Ultra Generate 001
- + Captures a professional poster layout with a centered focal point.
- + Used the requested color palette effectively across the design.
- − Failed to follow the requested 6-step chronological structure, opting for a circular layout with nonsensical steps.
- − Text is mostly gibberish or repetitive, failing to provide the requested information labels.
- − Icons do not clearly correspond to the specific mission phases requested (e.g., Saturn V, lunar orbit).
Verdict: GPT Image 1.5 is the clear winner as it followed every instruction, including the specific 6-step sequence, icon types, and text labels. Imagen 4.0 Ultra generated a visually appealing layout but failed significantly on prompt adherence, producing garbled text and ignoring the specific chronological steps requested.
GPT Image 1.5
OpenAI's state-of-the-art image generation model with better instruction following and adherence to prompts
Imagen 4.0 Ultra Generate 001
Google's Imagen 4.0 Ultra model offering the highest fidelity and resolution for professional-grade image generation