OpenAI's state-of-the-art image generation model with better instruction following and adherence to prompts
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
GPT Image 1.5
#7 of 62 in Text-to-Image
Recraft V4.1
#53 of 62 in Text-to-Image
Where the votes landed
GPT Image 1.5
0%
win rate
Ties
0%
Recraft V4.1
0%
win rate
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
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 realism in the glass and liquid-like reflections.
- + Accurate placement of all four spatial requirements.
- + Natural lighting that matches the 'soft window light' description.
- − The plant in the background is quite large, bordering on a bush rather than a small plant.
Recraft V4.1
- + Perfect geometric cube and sphere shapes.
- + Clear visibility of the plant through the glass as requested.
- + High-quality textures on the book and wooden table.
- − The sphere is floating in the center of the cube, which feels less grounded and realistic than Model A.
- − The reflection on the table doesn't perfectly align with the base of the cube.
Verdict: Both models followed the complex spatial prompt perfectly. GPT Image 1.5 is the preferred winner because it presents a more physically grounded scene where the sphere sits on the bottom of the cube, whereas Recraft V4.1 has the sphere levitating, which wasn't requested.
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 depiction of realistic skin textures and clothing details.
- + Highly convincing bike mechanics and accessories like the toolkit and rags.
- + Strong adherence to the focal length and depth of field requirements.
- − The car in the background lacks the requested motion blur, appearing mostly static.
- − Framing is a bit too clean and centered for the 'imperfect framing' prompt.
Recraft V4.1
- + Successfully captures the requested motion blur from passing cars.
- + Good sense of mood and cinematic lighting that matches the 'candid' prompt well.
- + Better execution of the 'imperfect framing' with a more dynamic, off-center composition.
- − The hands and the object he is holding are anatomically confused and blurry.
- − The bicycle structure is simplified and lacks the mechanical realism found in the other image.
- − Visual quality is grainier with less clarity in the subject's face.
Verdict: GPT Image 1.5 wins on technical execution and realism, providing a much more convincing subject with detailed skin and a logically constructed bicycle. Recraft V4.1 adhered better to secondary prompt details like motion blur and imperfect framing, but failed on the core quality of the human subject and his hands.
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
- + Excellent high-contrast warm lighting reflecting off the armor
- + Very clear and detailed braided hair with specific beads mentioned in the prompt
- + Highly realistic skin texture with grime and distinct scars
- − The bokeh sparks are a bit large and slightly distracting alongside the face
Recraft V4.1
- + Strong cinematic composition and emotive facial expression
- + Good rendering of the engraved patterns on the pauldrons
- + Accurate depiction of dirt and weathering on the skin
- − The torchlight effect on the metal is less pronounced/warm than Image A
- − The beads in the braids are less distinct compared to Image A
- − The armor texture lacks the worn, weathered feel of the character's face
Verdict: GPT Image 1.5 followed the prompt more closely in terms of lighting and specific details like the beads and the highly detailed texture on the leather and armor. While Recraft V4.1 has a strong cinematic feel, GPT Image 1.5 feels more 'battle-worn' and better captures the warm torchlight reflection requested.
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
- + Excellent typography with clean hierarchy and perfect spelling.
- + High-quality, appetizing food photography that is well-integrated.
- + Professional layout that mimics a real-world usable menu.
- − The grid for the food photos is a bit irregular in sizing compared to a strict geometric grid.
- − Some minor overlapping of the pizza crust with the section divider line.
Recraft V4.1
- + Strong minimalist aesthetic with bold, clear headers.
- + Accurate adherence to the grid-style photo layout.
- + Good use of whitespace for a clean look.
- − One photo is floating awkwardly out of alignment with the column.
- − The food photography is very small and less appetizing than the competition.
- − Slight text artifacts and inconsistent spacing in the item descriptions.
Verdict: GPT Image 1.5 is the clear winner as it produces a professional, market-ready menu with legible text, perfect spelling, and much higher quality food photography. While Recraft V4.1 followed the minimalist grid prompt well, it suffered from alignment issues and the food images felt like small, low-resolution placeholders rather than integral design elements.
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 adherence to the 'fiery, glowing effect' for all text elements.
- + Superior photographic texture and detail on the food components, especially the patty and bun.
- + Dynamic and high-energy composition that fills the frame effectively.
- − Small artifacts in the debris and sparks create a slightly busy visual.
- − The euro symbol in the starburst is slightly deformed.
Recraft V4.1
- + Extremely clear and stylistically accurate fiery typography for the main title.
- + Clean, professional-looking layout with clear separation of elements.
- + Highly legible price starburst.
- − The food components look somewhat plastic and lack the photorealistic texture requested.
- − The background feels more like static smoke rather than a 'fiery background with glowing embers'.
- − Missing some requested ingredients like the bottom bun.
Verdict: GPT Image 1.5 is the clear winner as it captured the specified 'fiery' atmosphere and photorealistic textures much more effectively than Recraft V4.1. While Recraft V4.1 produced very clean typography, its burger components lack the appetizing detail and dynamic energy required by the prompt.
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 throughout the board
- + Extremely consistent handwriting style that looks authentically human
- + Perfect spelling and logical completion of the truncated prompt item
- − The cursive for the title is quite stylized, making the 'A' in 'April' look slightly like an 'H'
Recraft V4.1
- + Strong contrast between the white chalk and dark board
- + Accurate spelling of all requested menu items
- + Good inclusion of background bokeh to suggest a café environment
- − The handwriting style is less consistent, with the title appearing more blocky than cursive
- − Lettering in 'Truffle' is slightly messy with overlapping strokes
- − The chalk texture is a bit too uniform, losing some of the 'smudge' realism found in Model A
Verdict: GPT Image 1.5 performed exceptionally well, capturing a very realistic chalk texture and organic handwriting consistency that truly feels handwritten rather than like a digital font. Recraft V4.1 provided a clear and accurate board, but the handwriting felt slightly more mechanical and less elegant in the title section compared to GPT Image 1.5.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
GPT Image 1.5
- + High level of textural detail in the space suit and horse's coat.
- + Cinematic composition with a solid ground plane that adds to the surreal scale.
- + Excellent lighting and environmental storytelling with the lunar lander and planets.
- − Failed the primary negative constraint by placing the astronaut on top of the horse.
- − The dusting effect on the moon surface feels slightly grounded rather than truly surreal.
Recraft V4.1
- + Vibrant, ethereal lighting and beautiful nebula effects in the background.
- + Good sense of motion and weightlessness with the horse floating among asteroids.
- + Clean, sharp rendering of the astronaut's gear.
- − Failed the primary negative constraint by placing the astronaut on top of the horse.
- − Anatomy of the horse's front legs appears slightly awkward or elongated.
Verdict: Both models completely failed the negative constraint to have the 'horse on top', instead providing the cliché interpretation of an astronaut riding a horse. GPT Image 1.5 is the preferred image due to its superior cinematic composition, grounded lunar environment, and higher level of intricate detail compared to Recraft V4.1.
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 fur texture and lighting on the capybara.
- + Successfully captures the bored, indifferent expression of the human traveler.
- + High-quality photorealistic rendering of the taxi interior and clothing textiles.
- − The capybara's paws look slightly claw-like and unnatural on the steering wheel.
- − The framing is a bit tight, losing some of the Manhattan background.
Recraft V4.1
- + Natural side-view composition that shows more of the car's interior and the city lights.
- + Accurately places the human passenger in a detached, realistic pose.
- + The capybara's interaction with the steering wheel feels anatomically plausible for a surreal image.
- − The capybara is not wearing the requested dark jacket.
- − The human passenger's hands and phone interaction have minor anatomical warping.
- − The bokeh in the background is a bit less distinct than image A.
Verdict: GPT Image 1.5 wins due to superior detail and strict adherence to the prompt, particularly including the dark jacket which Recraft V4.1 missed. While Recraft V4.1 offers a better wide-angle composition, GPT Image 1.5's photorealism and character textures make the surreal scene feel more grounded and high-quality.
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 rendering with zero spelling errors.
- + Perfect adherence to style such as the thorn border and scroll banner.
- + Superior composition that balances the vintage parchment feel with cinematic lighting.
- − The parchment texture is very heavy, giving it a slightly cluttered feel.
Recraft V4.1
- + Strong gothic font choice for the main heading.
- + Includes additional thematic elements like skulls in the corners.
- + Good atmospheric depth in the background sky.
- − The text on the scroll banner is tiny and difficult to read.
- − Missing the explicit 'Date/Time/Location' labels requested in the prompt.
- − The main title text is cut off slightly on the left edge.
Verdict: GPT Image 1.5 is the clear winner as it perfectly captured every specific text requirement and formatting detail without errors. While Recraft V4.1 has a nice artistic style, it failed to render the sub-header text clearly and missed the specific labels for the event details.
Isometric Miniature Diorama Scenes
Text-to-Image“Create a clear, 45° top-down isometric miniature 3D cartoon scene of Japan's signature dish: sushi, with soft refined textures, realistic PBR materials, gentle lighting, on a small raised diorama base with minimal garnish and plate. Solid light blue background. At top-center: 'JAPAN' in large bold text, 'SUSHI' below it, small flag icon. Perfectly centered, ultra-clean, high-clarity, square format.”
AI Judge Analysis
GPT Image 1.5
- + Perfectly follows text instructions with bold, centered stacked typography.
- + Highly detailed isometric diorama with excellent PBR textures on wood and ceramics.
- + Superior visual richness with added thematic elements like the teapot and soy sauce.
- − The composition is slightly more cluttered than 'minimal' might suggest.
- − Chopsticks are slightly thick relative to the sushi scale.
Recraft V4.1
- + Clean, minimal aesthetic that adheres well to 'minimal garnish and plate'.
- + Soft, gentle lighting creates a more stylized cartoon feel.
- + Accurate isometric perspective and centered layout.
- − Failed to place 'SUSHI' below 'JAPAN', placing the flag in between instead.
- − Typography is thin and lacks the requested 'large bold' impact.
- − The texture of the fish is slightly less realistic compared to the PBR request.
Verdict: GPT Image 1.5 delivered a much more professional and high-fidelity 3D render, excelling in texture work and typography placement. While Recraft V4.1 adhered better to the 'minimal' aspect of the prompt, its failure to correctly stack the text and its less impressive material rendering makes GPT Image 1.5 the clear winner.
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 the 'big expressive eyes' and 'tumbling together' prompt elements
- + Very vibrant and joyful color palette
- + High-detail texture on the fur of each animal
- − The fox kit's paw on the right has some anatomical distortion/warping
- − Visual style feels slightly more digital/artificial than photorealistic
Recraft V4.1
- + Superior photorealistic lighting and backlighting (rim light) on the fur
- + Better dynamic composition showing the 'playfully chasing' action
- + More naturalistic anatomy and scale for the different animals
- − The bunny is missing its rear legs/lower body in the tall grass
- − The fox's eyes are less 'big and expressive' compared to the other animals
Verdict: GPT Image 1.5 creates a very cute, illustrative, and sentimental image that perfectly captures the 'big eyes' and 'tumbling' aspects of the prompt. However, Recraft V4.1 wins by providing a much more sophisticated level of photorealism, particularly in its handling of golden sunrise lighting and more natural-looking fur textures and animal interactions.
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 a professional custom-lettered feel.
- + Rich, appealing vintage color palette and texture.
- + Superior vector emblem style with high-quality shading on the cloche.
- − Failed the requirement for a light background by using a black background.
- − The steam element is a bit thick and stylized compared to the relative realism of the cloche.
Recraft V4.1
- + Perfectly adhered to the light background requirement.
- + Clean, minimalist composition that works well for a modern-vintage logo.
- + Accurate text rendering and correct placement of the banner.
- − The cloche is very small relative to the text, unbalancing the composition.
- − Typography feels a bit generic compared to the stylistic flair of the other model.
- − The steam lines are very thin and look slightly amateurish.
Verdict: GPT Image 1.5 produced a much higher quality graphic with professional-grade typography and texture, though it completely missed the prompt's instruction for a light background. Recraft V4.1 followed all instructions including the background color, but its composition is top-heavy with text and the graphical elements lack the same level of artistic polish. GPT Image 1.5 is preferred for its superior aesthetic execution despite the background error.
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
- + Excellent layout with logical panel flow that tells a clear story.
- + Matches the flat-vector illustration style perfectly with high-quality icons.
- + Includes accurate text and bonus content like the 'Tranquillity' landing site and astronaut silhouettes.
- − Slightly deviates from the specific 'NASA-inspired palette' by including green and brown on the Earth and Moon.
Recraft V4.1
- + Strict adherence to the requested NASA-inspired color palette (navy, white, muted red).
- + Clean, minimalist iconography that fits the 'modern vector' description.
- + Correct sequential numbering for the mission steps.
- − Step 5 is incorrectly numbered as '04'.
- − The icons for Descent and Landing are identical, failing to show the change in state requested in the prompt.
- − Included 'Saturn Vicon' as text, seemingly misinterpreting a instruction to use a Saturn V icon.
Verdict: GPT Image 1.5 is the clear winner because it successfully illustrates the progression of the mission through distinct icons for each stage, whereas Recraft V4.1 provides identical illustrations for the final two steps. GPT Image 1.5 also demonstrates superior layout and artistic execution, making for a much more effective infographic despite minor palette shifts.
Explore each model
Recraft's V4.1 standard tier text-to-image model — refines V4's photorealism with more natural lighting, softer gradients, and sharper illustration styles for everyday creative work