Fast distilled version of Black Forest Labs' FLUX.2 [dev] optimized for speed and cost efficiency.
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev] Flash
#5 of 62 in Text-to-Image
Imagen 4.0 Ultra Generate 001
#33 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [dev] Flash
60.0%
win rate
Ties
10.0%
Imagen 4.0 Ultra Generate 001
30.0%
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 [dev] Flash
- + Perfect adherence to physics with the sphere resting on the bottom of the cube
- + Realistic glass texture with dust and reflections
- + Highly photorealistic table and plant details
- − The book is slightly oversized relative to the cube
Imagen 4.0 Ultra Generate 001
- + Clean, sharp aesthetic with legible text on the book
- + Strong composition and lighting
- − Physically impossible internal sphere that appears to be floating
- − The plant is mostly beside the cube rather than behind it as requested
Verdict: FLUX.2 [dev] Flash produced a much more realistic and physically grounded image where the sphere correctly rests on the bottom of the glass container. While Imagen 4.0 Ultra is visually striking and features impressive text rendering, it failed the spatial logic of the prompt by making the sphere float and placing the plant primarily beside the 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.”
AI Judge Analysis
FLUX.2 [dev] Flash
- + Excellent motion blur on background vehicles which adds to the street photography realism.
- + Complex bicycle structure is handled surprisingly well with tools on the ground.
- + Very effective wet pavement reflections and rainy atmosphere.
- − The man's hands have structural issues where they meet the bicycle handles.
- − The front wheel of the bicycle is slightly disconnected from the fork.
Imagen 4.0 Ultra Generate 001
- + Incredible skin texture and facial detail, looking very realistic and non-stylized.
- + Clothing texture and the way rain sits on the jacket is highly convincing.
- + Excellent shallow depth of field and framing that feels like a genuine candid shot.
- − Lack of motion blur on the passing cars, which was a specific prompt requirement.
- − Slight anatomical issues on the primary hand holding the tool.
Verdict: FLUX.2 [dev] Flash followed the prompt more closely by including the requested motion blur, and it better captured the atmosphere of a rainy street. However, Imagen 4.0 Ultra Generate 001 produced a significantly more realistic human subject with superior skin textures and convincing wet-weather clothing, even though it missed the specific detail about moving cars.
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 [dev] Flash
- + Excellent depiction of surface textures on leather and cloth.
- + More naturalistic lighting with soft torch glow.
- + Highly lifelike eyes and skin pores.
- − The 'faint scars' look more like fresh bloody cuts.
- − The beads on the braids are a bit small and less distinct than Model B.
Imagen 4.0 Ultra Generate 001
- + Superb braided hair detail with clear, silver beads.
- + Strong, cinematic lighting effects with high-contrast shadows.
- + More accurate interpretation of 'faint scars' on the face.
- − The armor texture looks slightly more 'CG' and less like real metal compared to Model A.
- − The facial skin texture is a bit too smooth for a battle-worn character.
Verdict: FLUX.2 [dev] Flash produces a more grounded and realistic portrait with superior texture work on the materials, while Imagen 4.0 Ultra Generate 001 offers a more stylized, cinematic look with impressive hair and bead details. FLUX.2 is the winner for its more convincing 'lifelike' eyes and the complex interplay of light on the weathered metal armor.
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 [dev] Flash
- + Excellent typography rendering for titles and price points.
- + Strong adherence to the 'colorful food photos in grid' with vibrant accent borders.
- + Professional layout that feels like a real single-page casual dining menu.
- − The placeholder text for item descriptions is garbled.
- − The food photos lack variety, showing almost exclusively pizza even in the 'Appetizers' section.
Imagen 4.0 Ultra Generate 001
- + Provides a much better variety of food photos corresponding to different categories.
- + Clean, hyper-minimalist aesthetic that matches the prompt.
- + Logical grouping of sections with clear headings.
- − The text is highly illegible and lacks consistent font weights.
- − The layout has excessive empty white space at the bottom, making it feel unfinished.
- − Missing the 'vibrant accents' requested in the prompt beyond small color swatches.
Verdict: FLUX.2 [dev] Flash produced a much more finished and professional-looking menu with superior text rendering and bold colorful accents, despite the repetitive pizza imagery. Imagen 4.0 Ultra provided better food variety but failed on typographical clarity and composition, leaving the bottom half of the design empty.
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 [dev] Flash
- + Excellent typographic rendering with a realistic fire texture.
- + The lighting on the burger components matches the fiery environment perfectly.
- + The exploded view feels very natural and dynamic with realistic sauce droplets.
- − The 'starburst' shape for the price is a bit soft and lacks sharp defined points.
Imagen 4.0 Ultra Generate 001
- + Highly vibrant colors and a dramatic, swirling background that enhances the sense of energy.
- + Very sharp and clean text rendering for all requested labels.
- + Good separation between layers of the burger.
- − The cheese lacks the realistic texture and melt-look found in the other image.
- − The lighting on the burger feels a bit more like studio lighting than environmental fire lighting.
Verdict: Both models followed the prompt exceptionally well, including all text elements and the specific price. FLUX.2 [dev] Flash is the preferred choice because its lighting feels more integrated with the background, and the texture of the burger ingredients looks more photorealistic than Imagen 4.0 Ultra, which has a slightly more artificial sheen.
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 [dev] Flash
- + Excellent chalk texture with realistic grit and smudge marks
- + Followed the prompt instructions for natural variations in letter size and slant
- + Beautifully integrated environmental lighting from the café setting
- − Repeated the price and the word 'Cookies' awkwardly in the third item
Imagen 4.0 Ultra Generate 001
- + Clean and highly legible text following all requested menu names
- + Perfect spelling and consistent layout of pricing
- − The text looks more like a digital font than genuine hand-applied chalk
- − The chalk texture is too uniform and lacks the realistic pressure-sensitive strokes seen in the other model
Verdict: FLUX.2 [dev] Flash produces a much more authentic 'handwritten' result with a gritty, realistic chalk texture and natural environmental integration, even though it stuttered on the third menu item's text. Imagen 4.0 Ultra is technically perfect in its text rendering and layout, but the result looks like a digital overlay or a printed vinyl decal rather than a hand-drawn chalkboard.
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 [dev] Flash
- + Excellent cinematic lighting and depth of field
- + High level of texture detail in the astronaut suit and horse mane
- + Dynamic, floating composition that feels like open space
- − Completely failed the negative constraint to have the horse on top
- − Anatomically awkward horse limbs in the floating posture
Imagen 4.0 Ultra Generate 001
- + Creative addition of space-themed gear for the horse including a helmet and light-up boots
- + Polished, clean aesthetic with sharp rendering
- + Clear, well-defined astronomical elements like the ringed asteroid
- − Completely failed the negative constraint to have the horse on top
- − Static composition compared to the cinematic feel of the first image
Verdict: Both FLUX.2 [dev] Flash and Imagen 3 Ultra failed to follow the specific spatial instruction for a 'horse on top' of the astronaut, both defaulting to the standard horse-riding image. FLUX.2 [dev] Flash is the better artistic rendition due to its superior cinematic lighting and atmospheric depth, whereas Imagen 3 Ultra looks more like a clean digital illustration.
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 [dev] Flash
- + Excellent photorealism with realistic skin and fur textures.
- + Strong cinematic composition through the windshield.
- + Natural-looking hands and phone grip for the passenger.
- − The passenger is technically in the passenger seat next to the driver, not the back seat.
Imagen 4.0 Ultra Generate 001
- + Follows the spatial prompt more accurately with the passenger in the back seat.
- + Very clean rendering and correct text on the taxi cap.
- + Clearly depicts both paws on the steering wheel as requested.
- − Visual style leans slightly towards 3D render/digital art rather than photorealism.
- − The capybara's claws look unnaturally like thick fingers.
Verdict: FLUX.2 [dev] Flash produces a much more convincing photorealistic image with cinematic lighting, though it fails to place the passenger in the back seat. Imagen 4.0 Ultra Generate 001 follows the positional instructions perfectly and includes all details like the specific driver's cap, but it lacks the gritty realism of the first image. FLUX.2 is the likely winner for its superior visual quality and lighting.
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 [dev] Flash
- + Excellent moody and cinematic lighting that fits the 'gothic' prompt.
- + High-quality, gritty texture on the parchment and background elements.
- + Complex and well-executed border design including webs and thorns.
- − Includes some hallucinated text characters in the event details section.
- − The scroll banner is slightly small and positioned lower than typical for that layout.
Imagen 4.0 Ultra Generate 001
- + Perfectly legible and accurate text for all requested details.
- + Clean, polished illustration style with high contrast.
- + Effective use of the moon to highlight the central composition.
- − Leans more toward a modern cartoon style rather than the requested 'vintage gothic' look.
- − The thorns in the border look a bit repetitive and digital.
Verdict: FLUX.2 [dev] Flash captures the requested 'vintage gothic' and 'cinematic' atmosphere much better with its dark, textured aesthetic, though it has minor text artifacts. Imagen 4.0 Ultra Generate 001 provides perfect text rendering and a very polished image, but its style feels more like a modern digital illustration than a vintage gothic poster.
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 [dev] Flash
- + Expertly renders PBR materials with realistic textures on the fish and wooden base.
- + Highly professional lighting and shadows that enhance the 3D diorama effect.
- + Excellent text rendering with clean, bold typography and a perfectly placed flag icon.
- − The sushi construction is slightly illogical, combining nigiri-style fish toppings on top of sliced maki rolls.
Imagen 4.0 Ultra Generate 001
- + Better variety of sushi types (tamago, ebi, salmon, maki, gunkan) which fits the 'miniature scene' request.
- + Perfectly hits the '3D cartoon' aesthetic with stylized, soft shapes.
- + Clean layout and accurate text and flag rendering.
- − Materials look more like plastic/clay than the requested 'realistic PBR materials'.
- − Lighting is a bit flat compared to the depth shown in the other model.
Verdict: FLUX.2 [dev] Flash produces a more sophisticated image with superior material rendering and lighting, making it look like a high-end 3D render. However, Imagen 4.0 Ultra Generate 001 provides a much better variety of sushi and a more charming 'cartoon' interpretation that better fits the 'miniature scene' prompt. FLUX.2 wins by a narrow margin due to the significantly higher quality of its textures and professional lighting.
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 [dev] Flash
- + Excellent adherence to the 'hyper-photorealistic' instruction with realistic fur textures and anatomy.
- + Superior lighting and atmosphere, featuring subtle god rays and highly realistic dew drops in the grass.
- + Natural composition with well-integrated animals that feel like they are inhabiting the same physical space.
- − Included an extra animal (a second bunny) not requested in the prompt.
- − The kitten's pose is slightly awkward with its front paws.
Imagen 4.0 Ultra Generate 001
- + Strict adherence to the requested animal count (one of each species).
- + Dynamic and playful posing that captures the 'joyful' vibe well.
- + Vibrant colors and clear butterfly interactions.
- − Fails the 'hyper-photorealistic' requirement, appearing more like a digital painting or 3D render.
- − The lighting and god rays feel artificial and overly 'glowy'.
- − Anatomy issues, specifically the fox's hands/paws looking more like primate hands than a kit's paws.
Verdict: FLUX.2 [dev] Flash significantly outperforms the competition in terms of realism, lighting, and texture, delivering a truly photorealistic image as requested. While Imagen 4.0 followed the specific animal count more accurately, its output has a distinctively 'AI-generated' illustrative look with anatomical errors on the fox, whereas FLUX.2 looks like a high-quality photograph.
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 [dev] Flash
- + Excellent typography adherence including the accent in 'Caffè'
- + Strong decorative texture that enhances the vintage aesthetic
- + Classic banner composition that feels cohesive and professional
- − The steam effect is a bit chunky compared to the fine lines of the logo
Imagen 4.0 Ultra Generate 001
- + Clean vector-style execution with precise line work
- + Accurate spelling and inclusion of all requested elements
- + Strong minimalist interpretation of the cloche dome
- − The 'Est. 1720' text is slightly off-center within its banner
- − Less texture than requested in the prompt
Verdict: Both models performed exceptionally well, following all text and stylistic requirements accurately. FLUX.2 [dev] Flash is the preferred choice as its composition feels more integrated and the 'vintage' texture is more successfully realized, whereas Imagen 4.0 Ultra feels a bit more modern and sterile by comparison.
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 [dev] Flash
- + Successfully followed the 6-step chronological sequence requested in the prompt.
- + High-quality rendering of specific icons like the Saturn V and Lunar Module.
- + Excellent text legibility for the main labels and names.
- − Repeats several labels (Launch/Landing) and icons unnecessarily.
- − The 'Sataurri' Iccòn' text is a minor hallucination of the prompt's descriptive text.
Imagen 4.0 Ultra Generate 001
- + Very clean, professional layout that feels like a real infographic poster.
- + Strong adherence to the 'NASA-inspired palette' with a white background and crisp vector look.
- + Consistent iconography style across the central hub.
- − Technically fails the prompt's logical sequence, replacing the requested 6 steps with generic 'lorem ipsum' style blocks.
- − Text is largely gibberish (MIOLIO, MISES11, MASED).
- − Does not show the specific requested icons in a way that relates to the mission steps.
Verdict: FLUX.2 [dev] Flash is the clear winner for its superior prompt adherence, correctly depicting the specific requested stages of the Apollo 11 mission from launch to landing with recognizable icons and legible text. While Imagen 4.0 Ultra Generate 001 creates a more aesthetically polished professional layout, it fails to follow the content instructions, filling the poster with gibberish and disconnected icons.
Explore each model
Google's Imagen 4.0 Ultra model offering the highest fidelity and resolution for professional-grade image generation