Black Forest Labs' compact, open-source image generation model with sub-second inference, optimized for production and near real-time applications with multi-reference support
Settled by community votes across 20 shared challenges, with an AI judge weighing in on each.
FLUX.2 [klein] 4B
#32 of 62 in Text-to-Image
GPT Image 1.5
#5 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [klein] 4B
0.0%
win rate
Ties
0.0%
GPT Image 1.5
100.0%
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
FLUX.2 [klein] 4B
- + Excellent shallow depth of field, giving a realistic photographic look.
- + Superior rendering of materials, especially the book cover and wooden table grain.
- + Naturalistic lighting from the left following the prompt precisely.
- − The glass cube has physically impossible interior reflections at the base.
GPT Image 1.5
- + Perfect adherence to all prompt elements including placement and color.
- + Clear visibility of the plant through the glass panels.
- + Higher resolution feel with sharp details on the book edges.
- − The lighting feels a bit flat and less directional compared to the other model.
- − The base of the cube looks slightly disconnected from the wood surface.
Verdict: Both models followed the complex spatial prompt perfectly, including the nested and layered objects. FLUX.2 [klein] 4B is the winner due to its significantly more realistic photographic quality and better handling of soft window light, whereas GPT Image 1.5 feels slightly more like a digital render.
Man and Car in California
Editing“Make a photo of the man driving the car down the California coastline”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Successfully merges both source images while preserving the specific Rolls Royce model.
- + Captures the sense of motion with blurred wheels and road lines.
- + Maintains the man's hair style and clothing from the source image.
- − The man's facial features look slightly different than the source image.
- − The lighting on the man's face doesn't perfectly match the bright exterior scene.
GPT Image 1.5
- + Excellent preservation of the man's facial features and joyful expression.
- + Highly realistic integration of the car, driver, and the specific Pacific Coast Highway aesthetic.
- + Lighting on the subject is very natural and consistent with the environment.
- − The car is cropped quite closely, showing less of the vehicle compared to the source.
- − The steering wheel appears a bit thick and simplified in detail.
Verdict: Both models handled the complex task of merging two distinct source images into a new environment very well. FLUX.2 [klein] 4B shows more of the car and conveys a better sense of motion, but GPT Image 1.5 is the winner due to its superior preservation of the man's identity and facial expressions from the source image, combined with more natural lighting.
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
FLUX.2 [klein] 4B
- + Excellent depiction of heavy rain and wet reflections on asphalt
- + The 50mm shallow depth of field is well executed with bokeh in the background
- + Great cinematic lighting and composition
- − Anatomical and physical errors: the man's leg merges into the bike frame and the pedaling mechanism is nonsensical
- − Lack of 'motion blur' on passing cars as requested; they appear relatively static
GPT Image 1.5
- + Highly realistic skin texture and fabric details
- + More accurate 'repairing' action with tools visible on the ground
- + Effective 'imperfect framing' that feels like a genuine candid street photo
- − The bike's rear wheel and chain assembly are structurally messy
- − Motion blur on the background car is subtle, though more present than in model A
Verdict: GPT Image 1.5 is the winner as it captures the 'street photography' aesthetic much better, providing a more believable skin texture and a realistic squatting pose for someone repairing a bike. While FLUX.2 [klein] 4B has beautiful lighting, it suffers from major AI artifacts where the subject's anatomy blends into the bicycle frame.
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
FLUX.2 [klein] 4B
- + Excellent detail on the engraved metal plate armor
- + Follows all prompt elements including braids with beads
- + Clear and symmetrical composition
- − Skin texture looks slightly plasticky compared to the other model
- − The lighting feels a bit artificial and flat on the face
GPT Image 1.5
- + Incredible skin texture with realistic dirt and scars
- + Superior use of lighting and bokeh to create a cinematic atmosphere
- + Very lifelike eye detail and natural-looking hair braids
- − The crop is slightly tighter, losing some of the shoulder armor detail
- − More chaotic composition compared to the centered approach of the competitor
Verdict: While FLUX.2 [klein] 4B followed the technical aspects of the prompt very well, GPT Image 1.5 produced a significantly more lifelike and cinematic image. GPT Image 1.5's mastery of skin texture, realistic grime, and the warm glow of torchlight creates a more convincing 'battle-worn' aesthetic.
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
FLUX.2 [klein] 4B
- + Features a high volume of high-quality food photography
- + Maintains a consistent, stylish modern minimalist aesthetic
- − Text is largely illegible gibberish
- − The grid layout feels slightly cluttered with too many overlapping sections
GPT Image 1.5
- + Excellent text legibility and logical menu descriptions
- + Highly accurate adherence to requested sections (Appetizers, Pizza, Mains)
- + Clean, professional layout that looks like a real-world document
- − Composition is a bit more traditional and less experimental than the other model
- − The image grid is functional but less dynamic than the first model's
Verdict: GPT Image 1.5 is the clear winner as it produces a functional, professional menu with perfectly legible text and logical pricing, whereas FLUX.2 [klein] 4B produces gibberish. GPT Image 1.5 also followed the specific section prompts perfectly, delivering a clean design suitable for actual use.
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
FLUX.2 [klein] 4B
- + Clean and polished graphic design aesthetic
- + Accurate rendering of the price in the starburst
- + Consistent lighting on the burger patty and bun
- − The main title text is completely garbled and misspelled
- − Failed the 'exploded' portion of the prompt as the burger is mostly assembled
- − The background is static rather than showing a sense of motion
GPT Image 1.5
- + Perfectly executed the 'exploded' burger concept with suspended components
- + Excellent text accuracy for the main title and secondary messages
- + Highly dynamic composition with great sense of motion and fiery effects
- − The starburst outline is a bit cluttered with fire particles
- − High contrast makes some of the vegetable details slightly crunchy/oversharpened
Verdict: GPT Image 1.5 is the clear winner as it followed all aspects of the prompt, including the complex 'exploded' layout and correct text spelling. FLUX.2 [klein] 4B failed to separate the burger components as requested and produced severe spelling errors in the prominent main title.
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
FLUX.2 [klein] 4B
- + Natural chalk smudge and dust textures on the board
- + Authentic variation in letter sizes and spacing
- + Excellent wooden frame detail
- − Numerous spelling errors including 'SPECIALS', 'Musheram', and 'Ootrpous'
- − Text is slightly less elegant and cursive than requested
GPT Image 1.5
- + Perfect spelling on every line including the complex menu items
- + Beautiful elegant cursive style that matches the prompt perfectly
- + Consistent chalk texture across all text
- − The lighting at the top of the frame is a bit washed out
- − The overall composition feels slightly more like a digital overlay than raw chalk
Verdict: GPT Image 1.5 is the clear winner as it correctly spelled every word in the prompt, whereas FLUX.2 [klein] 4B struggled significantly with basic spelling and letter formation. GPT Image 1.5 also followed the aesthetic request for 'elegant cursive' much more closely while maintaining high legibility.
Pose & Character Mashup
Editing“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Successfully replicates the exact dynamic body tilt and hair flow from Image 1.
- + Incorporates the scarf and sunglasses from Image 2 onto the new pose.
- − Failed to preserve the short hairstyle from Image 2, combining it with the long hair from Image 1 instead.
- − The facial features are heavily distorted and do not match the man in Image 2.
GPT Image 1.5
- + Perfectly preserves the facial features, skin tone, and hairstyle of the character in Image 2.
- + Accurately replicates the clothing details, including the specific graphic text on the sweatshirt.
- + Successfully adopts the crossed-leg pose and platform from Image 1.
- − Missed the extreme torso tilt and head orientation from Image 1, resulting in a more upright posture.
- − The right hand and left arm have slight anatomical clipping/merging issues at the edges.
Verdict: GPT Image 1.5 is the winner because it successfully maintains the character's identity, face, and clothing from Image 2 with high fidelity. While FLUX.2 [klein] 4B followed the extreme body tilt of Image 1 more closely, it failed the primary task of character preservation by giving the man long hair and a distorted face. GPT Image 1.5 effectively merged the character and the complex pose while keeping the person recognizable.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Excellent visual clarity and clean composition.
- + Good cinematic lighting and high resolution textures.
- − Failed the negative constraint to put the horse on top of the astronaut.
- − Very conventional interpretation of a common AI prompt.
GPT Image 1.5
- + More dynamic action with dust and movement effects.
- + Higher level of scene complexity including a lunar lander and planets.
- − Failed the negative constraint; the astronaut is still riding the horse.
- − Slightly cluttered composition with various celestial objects.
Verdict: Both models failed the specific spatial logic instruction to place the 'horse on top' of the astronaut, instead defaulting to the common trope of an astronaut riding a horse. FLUX.2 [klein] 4B produced a cleaner, more focused image, while GPT Image 1.5 offered a more detailed and busy background, but neither followed the surreal prompt instruction correctly.
Outfit Transfer Challenge
Editing“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Successfully preserved the subject's face, hair, and specific skin unique patterns (vitiligo).
- + Captured the complex plaid patterns and accessories like the bracelets and rings from current image 2.
- + Maintained the exact background lighting and beach setting.
- − The coat is missing the shirt layer beneath, leaving the chest exposed.
- − The placement of the hand in the pocket looks slightly distorted.
GPT Image 1.5
- + Excellent replication of the fabric textures of the navy peacoat and the plaid scarf.
- + Applied the clothing set including the black undershirt as requested.
- − Completely failed the 'keep face unchanged' instruction by cropping out the head.
- − The skin visible on the hands does not match the subject's unique vitiligo patterns from Image 1.
Verdict: FLUX.2 [klein] 4B follows the identity preservation instructions much better, keeping the subject's unique facial features and skin patterns intact while accurately transferring the clothing. GPT Image 1.5 failed the most basic part of the request by cropping out the subject's face entirely and failing to maintain the subject's skin details on the hands.
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
FLUX.2 [klein] 4B
- + Excellent photorealistic texture on the capybara's fur
- + Very clear and detailed passenger matching the 'bored businesswoman' prompt perfectly
- + Natural lighting and color depth
- − The car interior looks too small, resembling a compact car rather than a standard NYC Crown Vic or SUV taxi
- − The capybara's head has a slightly strange profile angle
GPT Image 1.5
- + The taxi driver cap is more accurate with the checkerboard pattern and 'TAXI' text
- + Superior composition with a more realistic wide taxi cabin perspective
- + The capybara's forward-facing expression is more 'professional' and human-like
- − Minor distortion in the human passenger's hands on the phone
- − Slightly more artificial 'HDR' look to the lighting compared to Image A
Verdict: Both models followed the prompt exceptionally well, but GPT Image 1.5 is the winner due to a more realistic vehicle composition and a much better accessory (the taxi cap). FLUX.2 [klein] 4B feels a bit cramped inside the car, although its rendering of the human passenger was slightly more detailed.
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
FLUX.2 [klein] 4B
- + Excellent character rendering with cinematic glowing lighting
- + Clean border with thorns and spiderwebs
- + Sharp focus on the central jack-o-lantern
- − Major spelling errors in the title text
- − Missing the 'Time' detail requested in the prompt
- − Incorrect date '206' and 'Locton' typo
GPT Image 1.5
- + Perfect text rendering for all requested information
- + Strong vintage gothic aesthetic with dark parchment textures
- + Complete adherence to all prompt details including the full scroll text and location
- − The jack-o-lantern rendering is slightly more textured/noisy than Model A
- − The full moon wasn't explicitly requested but adds to the composition
Verdict: While FLUX.2 [klein] 4B produces a very clean and visually appealing illustration, it fails significantly on text legibility and prompt adherence regarding event details. GPT Image 1.5 successfully follows every textual instruction with perfect spelling and maintains a consistent, high-quality vintage aesthetic suitable for an invitation.
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
FLUX.2 [klein] 4B
- + Excellent source preservation, keeping the facial features and clothing identical.
- + Realistic integration of long, wavy hair around the ears and forehead.
- − The hairline on the forehead looks slightly superimposed rather than growing naturally from the skin.
GPT Image 1.5
- + Natural, curly hair texture that fits the character's beard style well.
- + The hairline integration along the forehead is very seamless.
- − Slightly altered the shape of the glasses and the bridge of the nose.
- − The background landscape changed significantly compared to the original image.
Verdict: FLUX.2 [klein] 4B is the winner because it successfully added the hair while preserving all other elements of the original image, including the exact facial structure and background. Although GPT Image 1.5 produced a very natural-looking head of hair, it failed as an edit by regenerating the background and slightly shifting the likeness of the man.
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
FLUX.2 [klein] 4B
- + Excellent soft 3D cartoon textures that match the prompt's aesthetic
- + Very clean and minimal layout
- + Accurate 45° isometric perspective
- − Failed to render the full word 'SUSHI' (missing the 'I')
- − The flag icon is incorrect, resembling the flag of Austria instead of Japan
- − Lacks the requested 'small raised diorama base'
GPT Image 1.5
- + Perfectly adhered to all text and icon requirements, including the correct Japanese flag
- + Excellent interpretation of the 'raised diorama base' with layered textures
- + High visual quality with realistic PBR materials on the teapot and wood
- − Included significantly more items than the 'minimal garnish and plate' requested
- − The base extends a bit close to the edges of the square frame
Verdict: GPT Image 1.5 is the clear winner as it followed every complex instruction, including correct text rendering and the specific inclusion of a Japanese flag and raised diorama base. While FLUX.2 [klein] 4B captured the 'soft cartoon' texture well, it failed on the text and flag, and ignored the diorama base requirement.
Over-the-top cartoon caricature
Editing“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.”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Successfully captures the TV anchor environment with a professional desk and studio background.
- + Good caricature style that exaggerates facial features while remaining recognizable.
- + Includes multiple dogs as requested in a clear, illustrative style.
- − Completely misses the 'hockey' element of the prompt.
- − The text in the background is nonsensical and garbled.
GPT Image 1.5
- + Includes all three requested elements: TV anchor, dogs, and hockey.
- + The caricature of the subject maintains a strong likeness to the source image face.
- + Clever integration of hockey through the background screen and the dog wearing a helmet.
- − Hand holding the microphone has some anatomical issues with the finger placement.
- − The background is quite busy with multiple competing focus points.
Verdict: GPT Image 1.5 is the clear winner as it successfully incorporated all three thematic elements requested (TV anchor, dogs, and hockey), whereas FLUX.2 [klein] 4B completely omitted the hockey requirement. GPT Image 1.5 also exhibited a more polished rendering style and better likeness to the source subject.
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
FLUX.2 [klein] 4B
- + Excellent fur lighting and god ray composition
- + Clean, high-fidelity texture on the fox and puppy
- − Failed to include the baby bunny
- − Replaced the rabbit with a second kitten
GPT Image 1.5
- + Successfully included all four requested species
- + Captured the 'tumbling' and 'joyful' vibe more effectively
- + Great use of bokeh and dew sparkles
- − Anatomical issues with the kitten's paw reaching into the air
- − Some fur textures look slightly over-sharpened compared to natural photography
Verdict: While FLUX.2 [klein] 4B produces a very clean image with beautiful lighting, it failed the prompt adherence by omitting the baby bunny and adding a second kitten instead. GPT Image 1.5 successfully included the puppy, kitten, bunny, and fox kit while better capturing the dynamic tumbling motion requested, making it the superior choice for prompt accuracy.
Studio Ghibli Anime Style
Editing“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”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Excellent preservation of the original poses and character alignment.
- + Clear watercolor and ink-wash texture that feels very manual.
- + Maintains the distinct plaid pattern of the man's shirt effectively.
- − The woman on the right has lost her original angry expression, appearing content instead.
- − The background architectural details are a bit too busy compared to the typical Ghibli style.
GPT Image 1.5
- + Successfully captures the grumpy/angry expression of the woman on the right from the source image.
- + Warm, golden lighting and soft textures align perfectly with the 'dreamy' prompt requirement.
- + Captures the Ghibli character design style (large eyes, simple facial lines) more accurately.
- − The foreground woman is significantly blurred, losing some of the composition's impact.
- − Subtle artifacting on the textures makes it look a bit more like a digital filter than a hand-painted piece.
Verdict: Both models successfully interpreted the prompt, but GPT Image 1.5 is the winner for better preserving the emotional narrative of the original meme—specifically the woman's angry reaction—while hitting the stylistic Ghibli notes more accurately. FLUX.2 [klein] 4B provided a cleaner line-art style and preserved the poses better, but it completely changed the woman's expression to a smile, which misses the spirit of the source image.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.2 [klein] 4B
- + Natural-looking hair motion that follows a consistent wind direction.
- + Excellent source preservation including facial features and dog details.
- + Added leaves feel integrated with the lighting and environment.
- − The wind effect on the jacket and body is subtle compared to the hair.
GPT Image 1.5
- + Highly energetic hair movement that emphasizes the dynamic prompt.
- + Good preservation of the overall composition and subjects.
- − Several leaves appear to be 'floating' or pasted on without proper depth or blurring.
- − The hair flow is a bit chaotic and slightly alters the face shape near the hairline.
Verdict: Both models performed the edit well while maintaining the integrity of the original photo. FLUX.2 [klein] 4B is the winner because the added elements look more realistic and better integrated; the blowing hair follows a logical direction and the leaves feel part of the 3D space. GPT Image 1.5 achieved more 'energy' with the hair, but the leaves look more like a digital overlay than a natural part of the 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
FLUX.2 [klein] 4B
- + Clean vector aesthetic
- + Includes all requested elements like the 'Est. 1720' banner
- + Good texture on the background
- − Spelling error in the main text ('FLAXTION' instead of 'Florian')
- − Redundant text below the banner
- − The steam effect is very basic
GPT Image 1.5
- + Perfect text rendering of 'Caffè Florian'
- + Excellent vintage typography and illustrative style
- + Superior steam and shading detail on the cloche
- − Uses a solid black background instead of the requested 'light background'
- − Contrast is very high, moving away from the requested 'minimalist' feel
Verdict: GPT Image 1.5 is the winner because it successfully spelled the requested name 'Caffè Florian' and provided a high-quality illustrative style, whereas FLUX.2 [klein] 4B failed the text challenge with a significant misspelling. While GPT Image 1.5 failed on the background color, the superior typography and design detail make it much more useful as a logo concept.
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
FLUX.2 [klein] 4B
- + Clean vector aesthetic for the Saturn V illustration
- + Follows the muted color palette well
- − Text consists of severe gibberish in almost every label
- − Layout is cluttered and lacks logical flow for an infographic
- − Saturn V rocket is incorrectly landing on the moon surface
GPT Image 1.5
- + Excellent text rendering with correct spelling for all stage titles
- + Clear, logical grid layout that follows the requested 6-step sequence
- + Strong adherence to the flat-vector style with consistent iconography
- − The red color used is slightly more vibrant than a typical 'muted' NASA red
- − The profile silhouettes at the top are very basic compared to the detailed icons below
Verdict: GPT Image 1.5 is the clear winner as it successfully follows the structured 6-step prompt with accurate text, logical flow, and high-quality iconography. FLUX.2 [klein] 4B fails significantly on text legibility and infographic logic, resulting in a disorganized layout with nonsensical labels.
Explore each model
OpenAI's state-of-the-art image generation model with better instruction following and adherence to prompts