OpenAI's previous generation image model with higher quality than DALL-E 2 and support for larger resolutions
Settled by community votes across 9 shared challenges, with an AI judge weighing in on each.
DALL-E 3
18.5
arena score
#35 of 44 in Text-to-Image
Skill signature
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
FLUX.2 [max]
25.9
arena score
#11 of 44 in Text-to-Image
Vote tally
Where the votes landed
DALL-E 3
0.0%
win rate
Ties
0.0%
FLUX.2 [max]
100.0%
win rate
0.0%
0.0% ties
100.0%
Shared challenges
9
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Excellent photographic texture on the wooden surface and plant leaves
- + High artistic quality with a unique miniature-landscape interpretation inside the sphere
- + Strong contrast and dramatic lighting
- − Failed the spatial instructions: the book is inside the cube rather than on top of it
- − The 'glass cube' is rendered more as a wooden-framed case than a pure glass cube
FLUX.2 [max]
- + Perfect spatial adherence to the prompt's requested object placement
- + Highly realistic glass reflections and refraction
- + Accurately captures the 'soft window light from the left' instruction
- − Composition feels slightly clinical or like a 3D render rather than an artistic photograph
- − The secondary spheres reflected in the glass are slightly confusing
Verdict: FLUX.2 [max] followed the complex spatial instructions perfectly, accurately placing the red book on top of the cube and the blue sphere inside it. While DALL-E 3 produced a more visually striking and detailed image, it failed the core requirements of the prompt by inverting the positions of the objects.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Beautiful composition and reflection in the puddle.
- + Excellent cinematic lighting and atmosphere.
- + Good execution of the 'imperfect framing' request through the foreground bokeh.
- − Anatomical issues with the man's neck and bare feet which look distorted.
- − The man's skin and clothing texture look somewhat painterly rather than realistic.
- − The car in the background lacks the requested motion blur.
FLUX.2 [max]
- + Outstanding realism in skin texture and fabric.
- + Accurately captures the motion blur of the passing car as requested.
- + Highly realistic mechanical details on the bicycle and tools.
- − The composition is a bit more standard/centered than 'imperfect'.
- − Raindrops are subtle and less atmospheric than image A.
Verdict: FLUX.2 [max] significantly outperforms DALL-E 3 by delivering a truly realistic, photographic quality with natural skin textures and accurate motion blur. While DALL-E 3 creates a more artistic and cinematic composition, it suffers from anatomical distortions and a lack of the specific motion blur requested in the prompt.
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.”
DALL-E 3
FLUX.2 [max]
0% wins
0% ties
100% wins
AI Judge Analysis
DALL-E 3
- + Provides a layout of four variations in one view
- + Captures an artistic, high-contrast lifestyle food photography style
- + Effective use of blocking and white space for a premium feel
- − Text is completely illegible and garbled
- − Layout is more of a graphic design mock-up than a functional menu
- − Grid is inconsistent and busy
FLUX.2 [max]
- + Excellent text rendering with clear headers and pricing
- + Strong adherence to the grid-based food photo request
- + Clean, professional, and highly functional layout for a real restaurant
- − Some repetition in the pizza images
- − Text under headers is still nonsensical dummy text
Verdict: FLUX.2 [max] significantly outperforms DALL-E 3 by delivering a functional, readable menu design that follows all prompt instructions, including specific section headers. DALL-E 3 produces aesthetically pleasing graphic mockups, but the text is distorted and the layout is less practical for an actual casual dining menu.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Excellent sense of motion and dynamic energy
- + Beautifully integrated fiery environment and ground lighting effects
- + High level of texture detail in the burger components
- − Multiple spelling errors in the text, including 'MAGIC BURGR' and 'Limiited'
- − Failed to render the price in a starburst as requested
FLUX.2 [max]
- + Perfect text rendering with zero spelling errors
- + Accurately followed the prompt for a starburst price tag and fiery text effect
- + Clean and appetizing photorealistic food textures
- − The composition feels slightly more static compared to the explosive energy of the other model
- − The fiery background is a bit more generic in its execution
Verdict: While DALL-E 3 captures a more exciting and 'exploded' dynamic sense of motion, it fails significantly on the text requirements, with multiple spelling errors. FLUX.2 [max] provides a professional-grade advertisement with perfect text adherence, accurate starburst placement for the price, and excellent photorealism, making it the superior choice for a marketing use case.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Excellent texture on the capybara's fur and whiskers.
- + High-quality lighting consistency with shadows and highlights from the cab interior lights.
- + Clever background detail with a sign reading 'CAPYBARA' across the street.
- − Failed to include the businesswoman in the back seat.
- − The capybara's hands are quite long and slender, looking more like human-monkey hybrids than capybara paws.
FLUX.2 [max]
- + Successfully followed all instructions including the businesswoman on her phone in the back seat.
- + Accurate yellow taxi driver cap as requested.
- + Good composition that captures the requested 'bored' expression of the passenger.
- − Serious anatomical failure with the driver having human hands instead of capybara paws.
- − The taxi interior looks slightly less 'New York' and more generic than Model A.
- − Some lighting inconsistency on the passenger's face compared to the light sources.
Verdict: Both models failed to correctly render the capybara's paws—DALL-E 3 created strange hairy fingers while FLUX.1 [max] gave it fully human hands. However, FLUX.1 [max] is the preferred image because it adhered to the complex prompt's requirement for a businesswoman in the back seat, whereas DALL-E 3 completely ignored that subject.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Excellent vibrant colors and soft, appealing textures.
- + High level of visual clarity and 3D polish.
- + Clever integration of text and icons onto the diorama base.
- − Failed to place the text 'JAPAN' and 'SUSHI' at the top-center as requested.
- − The 'SUSHI' text is missing entirely.
- − The rice grains look more like spheres or bubbles than rice.
FLUX.2 [max]
- + Perfect adherence to text placement and content instructions.
- + Accurate representation of different sushi types (nigiri and maki).
- + Very clean, minimal, and professional isometric composition.
- − Lighting is a bit flatter compared to the vibrancy of Model A.
- − Wood texture on the base is slightly pixelated/blurry upon close inspection.
Verdict: Model B (FLUX.2 [max]) followed the complex layout instructions perfectly, placing the requested text and flag at the top-center while maintaining a clean isometric style. Model A (DALL-E 3) produced a more visually striking 3D render with better lighting and material feel, but it failed to follow the text prompts and basic positioning requirements.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Strong execution of the 'god rays' and warm golden lighting requested in the prompt.
- + Contains very expressive, large eyes that contribute to a high 'wholesome' vibe.
- + Vibrant color palette and high levels of detail in the fur texture.
- − The 'butterflies' are bizarre hybrids with bird/animal heads, which is a significant hallucination.
- − The image has a stylized, Pixar-like 3D render feel rather than the requested 'hyper-photorealistic' look.
- − The animals are oddly scaled relative to each other.
FLUX.2 [max]
- + Successfully achieves a hyper-photorealistic look with natural lighting and depth of field.
- + Accurately represents all four specific animals in the requested 'tumbling' and 'chasing' poses.
- + Superior source preservation of realistic butterfly anatomy and meadow physics.
- − The lighting is a bit more muted and less 'magical' than the god rays described in the prompt.
- − The baby fox has slightly unusual proportions in the front legs.
Verdict: While DALL-E 3 captures the magical lighting and cute expressions very well, its failure to generate anatomically correct butterflies (giving them furry animal heads) is a major flaw. FLUX.2 [max] provides a much more photorealistic and coherent scene that follows all parts of the prompt, including the specific movements of the animals, without strange artifacts.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Excellent emblem design with complex vector-style detailing.
- + Vibrant but classic color palette that fits the 'warm brown' request.
- + Includes the historical reference 'Est. 1720' accurately.
- − Failed to include the specific name 'Caffè Florian', replacing it with generic text.
- − Logo is a bit crowded for a 'minimalist' prompt.
FLUX.2 [max]
- + Perfect adherence to the requested text name 'Caffè Florian'.
- + Accurately represents the minimalist aesthetic with clean lines and subtle texture.
- + Expertly places the 'Est. 1720' on a banner as requested.
- − The steam effect is slightly faint compared to the other elements.
- − Composition is safe and standard for a circular logo.
Verdict: While DALL-E 3 produced a visually impressive emblem, it failed the most critical part of the prompt by hallucinating generic 'Coffee House' text instead of using 'Caffè Florian'. FLUX.2 [max] followed every instruction perfectly, including the specific name, the cloche dome, the banner, and the minimalist aesthetic.
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.”
DALL-E 3
FLUX.2 [max]
AI Judge Analysis
DALL-E 3
- + Excellent NASA-inspired color palette and vintage aesthetic.
- + High artistic complexity with a great sense of vertical layout.
- + Good use of iconography that feels professionally designed for a poster.
- − Failed the specific step-by-step instruction by generating generic layouts.
- − Text is mostly gibberish or decorative rather than readable data.
- − Includes incorrect spacecraft like the Space Shuttle instead of historical Apollo hardware.
FLUX.2 [max]
- + Perfectly followed the 6-step chronological sequence requested in the prompt.
- + Highly legible text with actual mission names and crew members.
- + Clean, modern vector style that matches the 'infographic' requirement.
- − Step labels at the top are repetitive/incorrect (uses 'Earth Orbit' three times).
- − Visual composition is a bit basic compared to a professional poster.
- − The astronaut icons look more like Soviet or corporate suits rather than NASA flight suits.
Verdict: While DALL-E 3 produced a more visually striking and artistic poster set, it failed significantly on technical accuracy and prompt adherence, including a Space Shuttle which is historically incorrect for Apollo 11. FLUX.2 [max] successfully mapped out the specific numbered steps and provided readable, relevant text, making it much more useful as an actual infographic despite its repetitive headers.
Next steps
Explore each model
Black Forest Labs' flagship image generation model delivering state-of-the-art quality with exceptional realism, precision, and consistency for both text-to-image and advanced image editing