Black Forest Labs' state-of-the-art image generation model with maximum quality and speed, supporting text-to-image and multi-reference image editing with up to 4MP output
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [pro]
26.3
arena score
#8 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.2 [pro]
100.0%
win rate
Ties
0.0%
Imagen 4.0 Ultra Generate 001
0.0%
win rate
100.0%
0.0% ties
0.0%
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.2 [pro]
Imagen 4.0 Ultra Generate 001
100% wins
0% ties
0% wins
AI Judge Analysis
FLUX.2 [pro]
- + Excellent photographic realism and soft lighting
- + Accurate glass refraction and reflections
- + Clean and simple composition that matches all prompt requirements
- − The sphere is resting on the bottom rather than being centered, though this is a valid interpretation
Imagen 4.0 Ultra Generate 001
- + Strong text rendering on the book spine
- + Rich colors and high contrast
- − The glass cube appears excessively thick or solid, creating distracting refraction artifacts
- − The blue sphere has a strange ghostly duplication/trail behind it
- − The perspective of the cube base seems slightly warped
Verdict: FLUX.2 [pro] produced a more believable and high-quality image with natural lighting and realistic glass physics. Imagen 4.0 Ultra Generate 001 struggled with the glass material, creating a confusing visual artifact behind the blue sphere and making the cube look like thick acrylic blocks rather than a single hollow or solid cube.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
100% wins
0% ties
0% wins
AI Judge Analysis
FLUX.2 [pro]
- + Excellent depiction of a realistic Japanese city street with believable puddles and background bokeh.
- + Very natural skin texture and age spots on the man's hands and face.
- + The bicycle mechanics (gears, chain, derailleur) are remarkably accurate and logical.
- − Minimal motion blur from passing cars, though the car in the background is slightly soft.
- − The hand positioning is slightly awkward where it meets the wheel hub.
Imagen 4.0 Ultra Generate 001
- + Successfully captured the requested 'imperfect framing' with a tighter, more candid crop.
- + Strong implementation of rain droplets on the man's jacket and skin.
- + Good shallow depth of field with visible raindrops catch in the light.
- − The bicycle structure is physically impossible, with handlebars protruding from the wall and a confusing wheel setup.
- − The man's right hand has anatomical issues, with fingers merging into the machinery.
- − Lacks the motion blur requested for passing cars.
Verdict: FLUX.2 [pro] is the superior model because it maintains high physical logic in complex objects like the bicycle and provides much more realistic skin textures. While Imagen 4.0 Ultra captured a better 'candid' mood through its lighting and framing, it suffered from significant anatomical and mechanical failures that broke the realism requested in the prompt.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent realism in skin texture and lighting.
- + Highly detailed engraving on the plate armor and aging on leather straps.
- + Dynamic composition with effective use of shallow depth of field and bokeh.
- − The beads in the hair are a bit more colorful/modern looking than expected for the setting.
- − The prominent forehead scar looks slightly like an overlay rather than deep skin damage.
Imagen 4.0 Ultra Generate 001
- + Symmetrical and clear armor design with sharp engraving details.
- + Consistent application of beads throughout the braided hair.
- + Good adherence to the 'warm torchlight' requirement with a visible torch source.
- − Skin texture appears slightly more waxy and less lifelike than Image A.
- − The scarring is represented by thin red lines that lack the 'battle-worn' depth requested.
- − Lighting on the face is a bit flat despite the presence of the torch.
Verdict: FLUX.2 [pro] is the winner due to its superior photographic realism, particularly in the rendering of skin, eyes, and the tactile quality of the armor and leather. While Imagen 4.0 Ultra Generate 001 provides a very clear and symmetrical composition, it has a more CG-like finish that lacks the gritty, atmospheric depth seen in FLUX.2 [pro].
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.2 [pro]
Imagen 4.0 Ultra Generate 001
100% wins
0% ties
0% wins
AI Judge Analysis
FLUX.2 [pro]
- + Clean, professional typography that uses bold sans-serif fonts effectively.
- + Logical dual-column layout that includes both images and a text-based price list.
- + High-quality, appetizing food photography that aligns well with the section headers.
- − Contains several spelling errors in headers like 'MINS' and item names.
- − The placement of grilled chicken under a 'Pizza' category showing photos of pizza is confusing.
Imagen 4.0 Ultra Generate 001
- + Excellent adherence to the 'grid' requirement with a very clean, structured layout.
- + Vibrant food photography that looks varied and high-end.
- + Very consistent minimalist aesthetic with plenty of white space.
- − The text is mostly gibberish or heavily distorted, making it unusable as a menu.
- − The overall composition feels a bit empty in the bottom right corner.
Verdict: FLUX.2 [pro] produces a more functional menu design with better-structured text and a realistic dual-column layout, despite some spelling errors. Imagen 4.0 Ultra Generate 001 creates a more visually striking grid of photos, but fails significantly on text legibility and content logic. FLUX.2 [pro] is preferred for its superior balance between design and readability.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent typography with a very realistic fiery, glowing effect.
- + Highly photorealistic food textures, especially on the meat patties and melting cheese.
- + Great sense of dynamic motion with sauce droplets and crumbs flying outwards.
- − The 'starburst' shape for the price is a bit muted and less distinct than requested.
Imagen 4.0 Ultra Generate 001
- + Strong composition with a swirling vortex background that enhances the 'magic' theme.
- + The price starburst is very well-defined with a clear fiery border.
- + Clean, bright rendering of the individual burger components.
- − The ingredients look slightly more illustrative and less photorealistic than Image A.
- − The 'LIMITED TIME ONLY' text is smaller and has less impact.
Verdict: Both models followed the prompt exceptionally well, but FLUX.2 [pro] stands out for its superior photorealism in the food textures and the high-quality execution of the fiery text. While Imagen 4.0 Ultra Generate 001 provides a more energetic background composition, FLUX's lighting and material details make for a more appetizing and professional-looking advertisement.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent chalk texture with realistic grit and uneven dispersal on the surface.
- + Perfect adherence to 'elegant cursive' request for the title.
- + Handwriting has natural slants and variations that feel human.
- − The board surface looks more like rough slate than a traditional chalkboard.
- − Some slight letter spacing issues in the bottom secondary text.
Imagen 4.0 Ultra Generate 001
- + Perfectly legible and organized text layout.
- + Clean, modern handwritten style that fits a café aesthetic.
- + High-quality rendering of the chalkboard surface and framing.
- − The text looks a bit like a digital chalkboard font due to very uniform stroke weights.
- − The title lacks the requested 'elegant cursive' style.
- − The chalk texture is very smooth and lacks the requested grainy, natural variation.
Verdict: FLUX.2 [pro] followed the prompt more accurately by providing a genuine cursive title and capturing the gritty, imperfect texture of real chalk. Imagen 4.0 Ultra produced a cleaner, more legible result, but failed to deliver the cursive handwriting requested for the title and the text appears slightly too uniform to be convincing as hand-drawn chalk.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Successfully placed the horse on top of the astronaut as requested by the specific instruction.
- + Excellent textures on the spacesuit and realistic lighting from the planet below.
- + Highly detailed background with multiple spiral galaxies contributes to a cinematic feel.
- − Anatomical glitch where the horse appears to be merging with the astronaut's back or a second horse body.
- − The surreal interpretation results in a slightly confusing middle-ground where the leg alignment is unclear.
Imagen 4.0 Ultra Generate 001
- + High visual clarity and creative details like the horse's space helmet and lit boots.
- + Good composition with a clear focal point and interesting celestial elements like the asteroid and crescent moon.
- − Failed to follow the specific spatial instruction for the horse to be on top of the astronaut.
- − Depicts a standard 'astronaut riding horse' trope which ignores the prompt's surreal reversal.
Verdict: FLUX.2 [pro] followed the difficult spatial instruction to place the horse on top of the astronaut, whereas Imagen 4.0 Ultra Generate 001 defaulted to a standard horse-riding-astronaut scene. Despite some anatomical confusion in its execution, FLUX.2 [pro] captures the surrealism and prompt intent much more effectively.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent photorealism with gritty, cinematic lighting that feels like a real New York taxi.
- + Precise adherence to the 'both front paws on the steering wheel' instruction with realistic capybara anatomy.
- + Authentic car interior details, including the dashboard UI and raindrops on the window.
- − The passenger is slightly out of focus, though this helps the cinematic feel.
Imagen 4.0 Ultra Generate 001
- + High clarity and well-lit characters.
- + Correct interpretation of the bored businesswoman passenger.
- − The capybara's paws look more like monster claws than natural rodent paws.
- − The perspective is awkward, placing the passenger almost level with the driver rather than in the back seat.
- − The overall image has a smooth, CG-like finish that lacks the requested photorealism.
Verdict: FLUX.2 [pro] is the clear winner as it perfectly captures the desired cinematic, photorealistic atmosphere of a New York night. While Imagen 4.0 Ultra creates a clean image, it fails on anatomical accuracy for the paws and lacks the realistic internal spatial depth of a car, whereas FLUX.2 [pro] feels like a genuine photograph.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent source parchment texture and weathered paper effects
- + Strong cinematic lighting with realistic atmospheric depth
- + Clean and legible typography that fits the gothic theme
- − The thorn border is a bit dark and blends into the background
Imagen 4.0 Ultra Generate 001
- + Highly detailed thorn and cobweb border design
- + Good adherence to all text elements in the hierarchy
- + Vibrant colors and clear contrast
- − The illustration style feels a bit more 'cartoonish' than 'vintage cinematic'
- − The scroll banner is slightly warped, affecting the text flow
Verdict: FLUX.2 [pro] wins by achieving a more authentic 'vintage gothic' aesthetic through its moody lighting and weathered parchment textures. While Imagen 4.0 Ultra has a more intricate border, its overall style feels more like a modern digital illustration compared to FLUX.2's cinematic poster feel.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent typography with a bold, modern feel that fits the diorama style.
- + Strong 3D isometric perspective and material rendering.
- + Great usage of a raised diorama base as requested.
- − The rice texture on the front nigiri is a bit lumpy and stylized compared to real sushi.
- − The flag icon is slightly smaller than what might be expected for the layout.
Imagen 4.0 Ultra Generate 001
- + Features a wider variety of realistic-looking sushi types.
- + Very clean, crisp rendering of materials like the shrimp and roe.
- + Accurate text and icon placement.
- − The text is plain and lacks the 'large bold' impact requested in the prompt.
- − The floating plate creates a less grounded 'diorama' feel compared to a base.
Verdict: FLUX.2 [pro] followed the prompt's layout and stylistic requirements more effectively, especially regarding the 'large bold' text and the 'raised diorama' structure. While Imagen 4.0 Ultra Generate 001 produced more detailed and varied sushi models, its typography was underwhelming and the plate lacked the physical grounding expected of a diorama scene.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Excellent photographic depth of field and lighting integration.
- + Highly realistic fur texture and dew drop details on the grass.
- + Natural interaction between the kitten and the fox kit.
- − Failed to include the baby bunny requested in the prompt.
- − The golden retriever's paw anatomy is slightly warped.
Imagen 4.0 Ultra Generate 001
- + Included all four requested animals: puppy, kitten, bunny, and fox.
- + Vibrant colors and strong 'god rays' lighting effects.
- + Great adherence to the 'butterflies' and 'meadow' elements.
- − The style leans towards '3D render' or 'illustration' rather than the requested 'hyper-photorealistic'.
- − The fox has an unusually long, bushy tail that looks slightly disconnected from its body.
Verdict: FLUX.2 [pro] produced a much more realistic and aesthetically pleasing image with superior lighting and textures, but it failed to include the bunny. Imagen 4.0 Ultra Generate 001 followed the prompt's subject list perfectly but resulted in a more artificial, digital art style that lacked the requested photorealism. Given the prompt's strong emphasis on 8K masterpiece realism, FLUX.2 [pro] is the stronger visual output despite the missing animal.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Perfect adherence to all text requirements including the 'grave accent' on Caffè.
- + Elegant circular composition creates a professional emblem feel.
- + Superior use of subtle paper-like texture on the background.
- − The shading on the cloche is slightly asymmetrical, though it fits the vintage style.
Imagen 4.0 Ultra Generate 001
- + Clean, minimalist line work that adheres well to the vector style.
- + Accurate text rendering and placement of elements.
- + Good use of warm brown and cream tones.
- − Composition feels a bit fragmented with the banner placed at the very bottom.
- − Less 'premium' feel than the emblem-style layout of the competitor.
Verdict: Both models followed the prompt exceptionally well, producing high-quality logos with correct spelling. FLUX.2 [pro] is the winner because its circular emblem composition is more cohesive as a brand identity, and the subtle background texture adds a level of sophistication that aligns perfectly with the 'vintage' request. While Imagen 4.0 Ultra is technically proficient, its layout feels more like a collection of icons rather than a finished logo design.
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.2 [pro]
Imagen 4.0 Ultra Generate 001
AI Judge Analysis
FLUX.2 [pro]
- + Strictly followed the sequential 6-step structure requested in the prompt.
- + Excellent flat-vector style with crisp lines and a professional infographic layout.
- + Very legible headings and clean, consistent iconography that matches the mission phases.
- − The body text beneath the headings is garbled 'lorem ipsum' style rather than English.
- − The trajectory line between Earth Orbit and Translunar is slightly disjointed.
Imagen 4.0 Ultra Generate 001
- + Includes a central NASA-style logo and a clean white background.
- + Maintains the requested color palette across several icons.
- − Completely failed the step-by-step instruction, instead creating a chaotic cluster of icons.
- − The icons are repetitive and do not clearly represent the specific mission phases requested.
- − Text rendering is poor, with many misspelled or nonsensical words ('MOSE', 'MASED').
Verdict: FLUX.2 [pro] followed the prompt's logical progression and stylistic requirements perfectly, producing a clear, chronological infographic of the Apollo 11 mission steps. In contrast, Imagen 4.0 Ultra provided a cluttered and disorganized layout that ignored the specific 6-step sequence and suffered from significantly worse text rendering.
Next steps
Explore each model
Google's Imagen 4.0 Ultra model offering the highest fidelity and resolution for professional-grade image generation