Black Forest Labs' 12-billion parameter multimodal flow transformer for in-context image generation and editing with character consistency, typography handling, and commercial-ready quality
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [pro]
20.3
arena score
#41 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Imagen 4.0 Ultra Generate 001
22.1
arena score
#34 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.1 Kontext [pro]
0%
win rate
Ties
0%
Imagen 4.0 Ultra Generate 001
0%
win rate
Shared challenges
13
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent adherence to the 'partially visible through glass' instruction for the plant
- + Highly realistic textures on the sphere and wooden table
- + Accurate thin-walled glass box construction
Imagen 4.0 Ultra Generate 001
- + Beautiful cinematic lighting with strong contrast
- + Excellent text rendering on the book spine
- + Sophisticated glass refraction effects
- − The plant is behind the cube but not visible through the glass as requested
- − The blue sphere appears to be floating unnaturally without a shadow inside the cube
Verdict: FLUX.1 Kontext followed the spatial instructions more accurately, specifically ensuring the green plant was visible through the glass cube. While Imagen 4.0 created a more visually striking image with impressive text on the book, it failed to render the plant through the glass and the sphere lacks the grounded realism found in the FLUX output.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent skin texture and realistic face for an elderly subject.
- + Perfectly captures the 'candid' and 'imperfect framing' request with a centered, slightly high-angle shot.
- + Very convincing rain effects and bokeh in the background.
- − The man appears to be riding or pushing the bike rather than 'repairing' it.
- − Misses the motion blur element requested for the passing cars.
Imagen 4.0 Ultra Generate 001
- + Accurately depicts the act of 'repairing' by showing the man working on the bike chain.
- + Good adherence to the 'reflections on wet pavement' and 'red bicycle' prompts.
- + Excellent depth of field and sharpness on the subject.
- − Static background lacks the requested 'motion blur from passing cars'.
- − The bicycle geometry is slightly warped near the pedals and chain area.
- − Lighting on the face feels a bit more studio-lit than 'natural/candid' street photography.
Verdict: Both models failed to include the requested motion blur for passing cars. FLUX.1 Kontext [pro] creates a much more realistic skin and face texture that feels like a true candid street photo, whereas Imagen 4.0 Ultra Generate 001 provides a better interpretation of the 'repairing' action. FLUX.1 Kontext [pro] is slightly preferred for its superior photographic realism and mood, despite the man's pose being more ambiguous.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Natural skin texture and subtle, realistic scarring.
- + Excellent engraving detail on the pauldrons with believable metallic sheen.
- + High-quality soft lighting and shallow depth of field.
- − The 'braids with small beads' requirement is very minimal, only appearing on one side.
- − The sparks are quite blurry and lack the 'bokeh' definition requested.
Imagen 4.0 Ultra Generate 001
- + Strict adherence to the 'braids with small beads' prompt with numerous visible beads.
- + Includes a visible torch as the light source, enhancing the theme.
- + More pronounced battle scars and dirt on the face.
- − The plate armor design is somewhat chaotic and overlaps the hair unnaturally.
- − The facial features feel slightly more processed/less organic compared to Model A.
- − The 'bokeh sparks' are numerous but look a bit like floating flat icons.
Verdict: FLUX.1 Kontext [pro] creates a more cinematic and photorealistic portrait with superior atmospheric lighting and texture. However, Imagen 4.0 Ultra Generate 001 provides much better adherence to the specific 'beaded braids' requirement and captures the 'battle-worn' aesthetic more aggressively. FLUX.1 Kontext [pro] is generally preferred for its visual sophistication, while Imagen 4.0 Ultra is better for literal prompt compliance.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent typography with clean, legible sections for Appetizers, Pizza, and Mains.
- + High-quality food photography with realistic lighting and appetizing textures.
- + Clean and balanced minimalist layout that feels like a real restaurant menu.
- − The text content is mostly gibberish/placeholder text.
- − The food shown under 'Mains' appears to be more pizza.
Imagen 4.0 Ultra Generate 001
- + Strict adherence to the 'grid' requirement with a structured image gallery.
- + Vibrant food photos that successfully capture a casual dining aesthetic.
- + Includes currency symbols which adds to the functional menu feel.
- − The typography is less professional with some distorted and overlapping characters.
- − The layout is somewhat repetitive and lacks the hierarchical 'bold sans-serif' title presence of the other model.
- − The grid feels cramped with many similar-looking small images.
Verdict: FLUX.1 Kontext [pro] captures the 'modern minimalist' aesthetic much more effectively with its professional use of whitespace and bold typography. While Imagen 4.0 Ultra follows the 'grid' instruction more literally, its text rendering and overall composition feel more cluttered and less like a professional design compared to the clean, realistic layout of FLUX.1 Kontext [pro].
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent typography rendering with a premium 3D glow effect.
- + Highly photorealistic textures on the bun and meat patty.
- + Atmospheric lighting with realistic fire and embers at the base.
- − Failed the 'exploded' instruction, as the burger components are mostly touching.
- − Redundant price tags clutter the composition.
Imagen 4.0 Ultra Generate 001
- + Perfect adherence to the 'exploded' requirement with clearly separated, floating layers.
- + Accurately included the requested starburst element for the price.
- + Dynamic composition with a circular motion blur effect in the background.
- − The lighting on the cheese and tomatoes feels slightly more illustrative than photorealistic.
- − Text is smaller and less impactful compared to the other model.
Verdict: Imagen 4.0 Ultra successfully captured the 'exploded' layout requested in the prompt, creating a much better sense of motion and individual components. While FLUX.1 Kontext produced superior photorealism and typography, it failed to separate the burger layers, resulting in a static image that ignored a core part of the creative brief.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent chalk texture showing individual particles and physical pressure.
- + Highly accurate spelling including a logical completion of the 'Brown But...' prompt.
- + The wood grain on the frame adds to the realism.
- − The 'Today's Specials' title is in print block letters rather than the requested 'elegant cursive'.
- − Minor typos in the bottom disclaimer text ('ous gluten tree').
Imagen 4.0 Ultra Generate 001
- + Followed the instruction for an elegant cursive title style more effectively.
- + Clean, perfectly legible text that completes the menu items correctly.
- + Good atmospheric lighting on the blackboard surface.
- − The text looks more like a digital marker or vector font than physical chalk on a board.
- − The letters lack the authentic grainy texture and varying opacity requested in the prompt.
Verdict: FLUX.1 Kontext [pro] offers a much more realistic interpretation of the 'chalk' texture requested, with convincing grain and handwriting styles. While Imagen 4.0 Ultra Generate 001 followed the 'cursive' instruction more closely and stayed cleaner, the result looks like a digital overlay rather than a physical chalkboard.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Perfect adherence to the difficult constraint of the horse being on top of the astronaut
- + Strong cinematic lighting with realistic textures on the spacesuit and horse fur
- + Excellent composition with a sense of scale and movement in space
- − Anatomical fusion where the astronaut's legs appear to end in horse hooves
- − Physical illogicality in strap placement for the riding gear
Imagen 4.0 Ultra Generate 001
- + High visual quality with vibrant colors and cosmic details
- + Great attention to detail via the horse's specialized space gear and glowing hooves
- − Complete failure to follow the core prompt instruction of 'horse on top'
- − Standard, non-surreal interpretation of the prompt
Verdict: FLUX.1 Kontext [pro] is the clear winner because it successfully followed the complex spatial instruction to place the horse on top of the astronaut, creating a truly surreal image. In contrast, Imagen 4.0 Ultra Generate 001 produced a standard astronaut-riding-a-horse image, ignoring the specific arrangement requested.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent photorealistic texture on the capybara's fur.
- + Natural lighting that matches a nighttime city environment.
- + Captures a candid, realistic angle from outside the vehicle.
- − The woman in the back is not clearly looking at a phone.
- − The composition is a bit tight, making it harder to see the full taxi interior.
Imagen 4.0 Ultra Generate 001
- + Perfect adherence to the prompt's character actions, including the woman looking at her phone.
- + Comprehensive view of the taxi interior and the city lights through the window.
- + Great character design for the capybara, including clear paw placement on the wheel.
- − The capybara's paws look more like sharp claws or human fingers in gloves, which is slightly anatomicaly incorrect.
- − The overall image has a slightly more 'digital' or CGI sheen compared to the photorealism of Model A.
Verdict: FLUX.1 Kontext [pro] creates a more convincingly photorealistic image with superior textures and lighting, but it misses the specific detail of the passenger looking at her phone. Imagen 4.0 Ultra Generate 001 provides a better compositional layout and matches every prompt instruction perfectly, despite having a slightly more synthetic visual 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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent typography style that fits the gothic theme perfectly.
- + The dark parchment texture and moody lighting create a sophisticated, vintage atmosphere.
- + Highly detailed and intricate spiderweb border.
- − Added a line of non-existent garbled text ('Your: Vorkleat: Iight & Spans') that wasn't in the prompt.
- − The 'vintage parchment' feel is a bit more like a modern poster than an aged document.
Imagen 4.0 Ultra Generate 001
- + Perfect text accuracy with no hallucinations or spelling errors.
- + Clearly incorporates all requested elements like thorns and the scroll banner with high visibility.
- + Clean, modern illustrative style with vibrant colors.
- − The aesthetic is more 'cartoonish' than the requested 'vintage gothic' and 'cinematic'.
- − The thorns and webs look like vector art rather than an integrated, organic part of a gothic scene.
Verdict: FLUX.1 Kontext [pro] creates a much more atmospheric and aesthetically pleasing image that captures the 'vintage' and 'moody' vibes of the prompt, but it fails by including a line of gibberish text. Imagen 4.0 Ultra Generate 001 provides a technically perfect execution of the text and specific objects, but the final art style is overly bright and clean, lacking the grit of a true gothic horror invitation. FLUX is preferred for visual quality, while Imagen is preferred for accurate information display.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent typography style that matches the 3D cartoon aesthetic.
- + Very clean rendered textures that look like high-quality clay or PBR material.
- + Perfectly centered composition with a clear diorama base.
- − The flag icon is stylized as a piece of paper or fabric rather than a standard icon.
- − Only features a single piece of sushi, which might be too minimal for 'Japan's signature dish'.
- − The rice grains are a bit oversized and uniform, bordering on bubbly textures.
Imagen 4.0 Ultra Generate 001
- + Includes a wider variety of sushi items as requested by 'signature dish'.
- + Highly accurate 45° isometric perspective.
- + Clean, professional typography and flag icon layout.
- − Floating plate effect feels slightly less grounded than Model A's diorama.
- − The white-on-white base and plate lacks contrast.
- − The garnish and sushi variety borders on exceeding the 'minimal garnish' constraint.
Verdict: Both models followed the prompt exceptionally well, but FLUX.1 Kontext [pro] captured the 'cartoon 3D' and 'soft refined textures' aesthetic more cohesively, specifically with its stylized 3D text. Imagen 4.0 Ultra Generate 001 provides a more comprehensive sushi spread and a more clinical isometric view, but the lack of contrast between the plate and the base makes the layout feel less like a 'diorama'.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent soft lighting and integration with the environment
- + Highly realistic fur texture and 'god rays' effects
- + Beautiful, natural color palette that feels cohesive
- − The 'tabby kitten' and 'bunny' look very similar and the kitten lacks distinct tabby markings
- − The animals are mostly sitting rather than 'playfully chasing' or 'tumbling'
Imagen 4.0 Ultra Generate 001
- + Successfully captures the 'playfully chasing' action through dynamic poses
- + Strong adherence to all animal types including a clear brown bunny and distinct tabby kitten
- + Vibrant colors and clear butterfly details
- − The art style leans toward a digital illustration rather than the requested 'hyper-photorealistic' look
- − The kitten's pose and anatomy look slightly awkward and flat
Verdict: FLUX.1 Kontext [pro] creates a much more realistic and visually stunning atmosphere with superior lighting and fur detail, but it fails to depict the animals actually playing or chasing. Imagen 4.0 Ultra Generate 001 captures the requested action and specific animal types much better, though it sacrifices photorealism for a more stylized, cartoonish aesthetic.
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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent typography with a hand-stamped, vintage finish
- + Superior texture that gives it an authentic aged paper feel
- + Bold composition with a clear, well-integrated banner
- − The 'steam' element is a bit thick and looks more like a leaf or abstract shape
Imagen 4.0 Ultra Generate 001
- + Elegant line work and sophisticated vector style
- + Accurate depiction of a cloche with thin, wispy steam
- + Clean and professional layout well-suited for a modern-retro brand
- − The accent on 'CAF FÈ' is slightly misaligned compared to model A
- − Lacks the 'subtle texture' requested in the background compared to the competitor
Verdict: FLUX.1 Kontext [pro] captures the 'vintage' and 'subtle texture' keywords much more effectively, producing a logo that looks like it was printed on physical cardstock. While Imagen 4.0 Ultra Generate 001 provides a cleaner, more precise vector illustration of the cloche and steam, FLUX.1's overall character and typography feel more authentic to the requested 1720 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.”
FLUX.1 Kontext [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Features a bold, attractive layout with a clear rocket illustration.
- + Matches the requested navy, white, and red color palette effectively.
- + Includes several of the requested labels like 'Translunar' and 'Landing'.
- − Confusing visual logic, showing Saturn with rings near the rocket top and mislabeling the Moon as Earth.
- − Fails to follow the 1-6 step sequence requested in the prompt.
- − Contains significant text artifacts and nonsensical labels like 'Gar Collins'.
Imagen 4.0 Ultra Generate 001
- + Accurately interprets the 'infographic poster' layout with a vertical step-by-step structure.
- + Strictly adheres to the requested color palette across all icons.
- + Better represents the individual mission components through consistent iconography.
- − Extremely poor text rendering with almost entirely gibberish words.
- − The center vertical graphic is abstract and doesn't clearly represent the flight path.
- − Icon choice for steps is somewhat repetitive and vague.
Verdict: Both models struggled significantly with the logic and text of the infographic. FLUX.1 Kontext [pro] produced a more visually striking single image but failed fundamentally on the educational content, mislabeling planets and ignoring the 6-step request. Imagen 4.0 Ultra Generate 001 followed the requested structure much better, presenting a clear vertical list of steps, but it rendered nearly all text as unreadable gibberish, making it less useful as a poster.
Next steps
Explore each model
Google's Imagen 4.0 Ultra model offering the highest fidelity and resolution for professional-grade image generation