Distilled version of Black Forest Labs' FLUX.2 [dev] outperforming it at a cheaper price. Developed by fal.ai.
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev] Turbo
27.8
arena score
#4 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Imagen 4.0 Fast Generate 001
17.6
arena score
#53 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.2 [dev] Turbo
66.7%
win rate
Ties
0.0%
Imagen 4.0 Fast Generate 001
33.3%
win rate
66.7%
0.0% ties
33.3%
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent photographic quality with realistic textures on the wood and book cover
- + Very accurate rendering of glass thickness and internal reflections
- + The plant looks realistically integrated and refracted through the glass
- − The blue sphere is slightly off-center and looks like it's floating slightly above the floor of the cube
Imagen 4.0 Fast Generate 001
- + Strong adherence to the spatial constraints of the prompt
- + Clean lighting with high-contrast shadows from the window
- + Well-defined geometric shapes for the cube and sphere
- − The glass cube lacks thickness and looks more like thin plastic sheets
- − The plant has a slightly artificial, CGI-like appearance compared to the wood texture
- − Unusual reflection of the blue sphere appears on the right side exterior of the cube
Verdict: FLUX.2 [dev] Turbo produces a more convincing photographic result with realistic material properties, particularly the weight and texture of the glass and the book. Imagen 4.0 captures the spatial elements well but suffers from a less realistic, slightly clinical aesthetic with thinner-looking glass and less natural plant textures.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent adherence to the 'elderly Japanese man' prompt with highly detailed skin and facial features.
- + Very realistic textures on the wet pavement, bicycle, and surrounding street environment.
- + Accurately captures all prompt elements including motion blur on passing cars and specific repair tools on the ground.
- − Minor anatomical/mechanical clipping where the man's leg appears to merge with or disappear behind the bike frame.
- − The rain effect is slightly static, looking more like a noise overlay than falling drops.
Imagen 4.0 Fast Generate 001
- + Strong use of 'imperfect framing' by looking through a window or doorway, creating a voyeuristic candid feel.
- + Beautiful bokeh and reflections on the wet asphalt.
- − Fails to show motion blur on the background car, which appears static despite the prompt request.
- − The man's face and upper body are partially cut off by the frame, losing some of the specified character detail.
- − The hands and the object he is holding are somewhat blurry and indistinct.
Verdict: FLUX.2 [dev] Turbo provided a much more complete and technically accurate response to the prompt, successfully incorporating the elderly man's features, the motion blur of traffic, and various repair tools. While Imagen 4.0 Fast Generate 001 had a creative take on 'imperfect framing', it failed to include the requested motion blur and sacrificed too much of the subject's detail. FLUX.2's image feels like a high-quality street photograph that captures the specific requested atmosphere perfectly.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
67% wins
0% ties
33% wins
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Perfect adherence to all prompt requirements including armor, braids, beads, and torchlight.
- + Excellent texture on the leather straps and engraved metal.
- + High visual quality with realistic skin tones, scars, and life-like eyes.
- − The torch in the background is a bit blurry, though this fits the shallow depth of field request.
Imagen 4.0 Fast Generate 001
- + The image of the man in the garden is high resolution and clear.
- − Completely failed to follow the prompt; shows an older man in a leather jacket in a garden instead of a paladin in armor.
- − Ignored all stylistic cues including torchlight, battle-worn details, and specific character features.
Verdict: FLUX.2 [dev] Turbo followed the prompt perfectly, delivering a highly detailed and cinematic portrait of a paladin with all requested elements like beads, engraved armor, and bokeh sparks. Imagen 4.0 Fast Generate 001 suffered a complete hallucination or failure, producing an image of an old man in a garden that bears no resemblance to the prompt.
Modern Clean Menu
Text-to-Image“Modern minimalist restaurant menu design, white background with colorful food photos in grid, sections for appetizers/pizza/mains, bold sans-serif fonts, vibrant accents, clean professional layout for casual dining.”
FLUX.2 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent typography with clean, bold sans-serif fonts
- + High-quality, professional food photography
- + Consistent modern design with bold color accents along the bottom
- − Nonsensical pricing ($250 for a pizza)
- − The grid layout of photos is a bit cluttered compared to the text area
Imagen 4.0 Fast Generate 001
- + Stronger adherence to the 'grid' layout requested in the prompt
- + Clean, organized multi-column text layout
- + Vibrant color blocks used as sectional accents
- − Spelling error in the main heading 'APETIERS'
- − Food photos lack variety, appearing to be mostly the same style of pizza
- − Text rendering is slightly less crisp than Model A
Verdict: FLUX.2 [dev] Turbo produces a more professional-looking graphic design with superior text rendering and high-end photography, though its pricing logic is flawed. Imagen 4.0 Fast follows the grid layout prompt more literally but suffers from a spelling error and repetitive imagery. FLUX is the preferred choice for a real-world design starting point despite the unrealistic prices.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent photorealistic texture on the meat patty and fresh vegetables.
- + Strong sense of motion with flying sauce droplets and embers.
- + Highly accurate adherence to the 'starburst' and 'fiery glowing text' requirements.
- − The sauce droplets are shaped a bit strangely, almost looking like worms.
Imagen 4.0 Fast Generate 001
- + Perfect text rendering with high legibility.
- + Clean composition with a clear focus on the food ingredients.
- + Accurate starburst price tag with glowing effect.
- − The patty texture looks slightly more plastic/artificial than Model A.
- − The 'exploded' effect is less dynamic and feels more like a static stack.
Verdict: FLUX.2 [dev] Turbo produced a more visceral and dynamic image with superior photorealistic textures on the food components and a greater sense of motion. While Imagen 4.0 Fast Generate 001 delivered cleaner text, FLUX.2 captured the 'fiery' atmosphere and 'exploded' burger prompt much more effectively.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent chalk texture with realistic smudges and varied line thickness
- + Perfect spelling and full adherence to the menu item names
- + Authentic cursive style for the title as requested
- − The layout for the octopus item is slightly cramped and spills onto a second line
Imagen 4.0 Fast Generate 001
- + Very clean and legible text alignment
- + Good overall framing of the chalkboard
- − Spelling errors present, such as 'Octuphus' and 'Cookes'
- − The text lacks realistic chalk texture, appearing more like a digital font
- − Contradictory duplicate lines for the 'Brown Butter' item
Verdict: FLUX.2 [dev] Turbo significantly outperformed Imagen 4.0 by providing a truly realistic chalk texture and accurate spelling across all requested menu items. While Imagen 4.0 produced cleaner lines, its text contained several typos and failed to capture the handwritten aesthetic as convincingly as FLUX.2.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.2 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent surface textures on the space suit and horse fur
- + Cinematic lighting with a clear horizon and atmospheric depth
- + Highly detailed background with planets and nebulae
- − Failed the specific spatial instruction for the horse to be on top of the astronaut
Imagen 4.0 Fast Generate 001
- + Clean, illustrative style with a smooth nebula background
- + Good anatomical structure for both the horse and the astronaut's gear
- − Failed the specific spatial instruction for the horse to be on top of the astronaut
- − Lower level of fine detail and texture compared to the competitor
Verdict: Both FLUX.2 [dev] Turbo and Imagen 4.0 failed the logic test inherent in the prompt, which specifically requested the horse be on top of the astronaut. Between the two standard interpretations, FLUX.2 [dev] Turbo is the superior image due to its much higher level of detail, realistic lighting, and cinematic composition.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent photorealism with cinematic lighting
- + Accurately places the passenger in the back seat as requested
- + Capture a gritty, authentic New York night atmosphere
- − The passenger is in the front passenger seat rather than the back seat
- − The capybara's paws look slightly human-like in structure
Imagen 4.0 Fast Generate 001
- + Features a very professional and detailed taxi driver uniform
- + High clarity and sharp focus on the capybara's fur texture
- + Correctly captures the bored expression of the businesswoman
- − The passenger is seated in the front passenger seat, failing the 'back seat' instruction
- − The steering wheel placement and interior perspective feel slightly distorted
Verdict: Both models failed to place the businesswoman in the back seat, instead placing her in the front passenger seat. FLUX.2 [dev] Turbo provides a much more convincing photorealistic aesthetic with lighting that feels grounded in a real environment, whereas Imagen 4.0 Fast Generate 001 feels slightly more like a digital render Despite the seating error, FLUX is the winner for its superior atmospheric 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.2 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent typography with perfect spelling in all requested text areas.
- + Superior atmosphere and texture, capturing the 'dark parchment' and 'gothic' look effectively.
- + Strong composition with a well-integrated border of thorns and webs.
Imagen 4.0 Fast Generate 001
- + Features a clean, high-contrast layout that is easy to read.
- + Accurately includes all specified elements like the scroll banner and twisted trees.
- − Contains spelling errors such as 'IINVIITATION' and 'NIGIT'.
- − The art style is a bit more 'cartoonish' rather than the requested 'cinematic' and 'vintage' gothic style.
- − The lighting lacks the moody depth found in the competitor's image.
Verdict: FLUX.2 [dev] Turbo followed the prompt perfectly, delivering flawless text and a rich, atmospheric gothic aesthetic. Imagen 4.0 Fast Generate 001 struggled with text rendering, introducing multiple typos, and produced a flatter, more generic vector illustration style that missed the 'vintage' and 'moody' requirements.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent PBR material rendering with realistic rice and salmon textures.
- + Perfect text rendering and layout for the typography and flag.
- + Highly detailed garnish like the ginger rose and wasabi peaks.
- − The perspective is slightly lower than the requested 45-degree isometric view.
Imagen 4.0 Fast Generate 001
- + Perfect 45-degree isometric perspective alignment.
- + Achieves a clean, stylized 'miniature 3D cartoon' aesthetic as requested.
- + Good centered composition with a multi-tiered diorama base.
- − The salmon texture looks overly plastic and less refined than the competition.
- − The text 'SUSHI' is slightly misaligned with the Japanese flag icon.
- − The inclusion of a white border around the square image was not requested.
Verdict: FLUX.2 [dev] Turbo produced a much higher quality image in terms of textures and materials, particularly with the hyper-realistic sushi rendering. While Imagen 4.0 Fast Generate 001 followed the isometric perspective more strictly, its lighting and materials feel flat and less professional compared to the refined output of FLUX.2.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Perfectly captures the 'playfully chasing' and 'tumbling' aspect of the prompt.
- + Excellent adherence to all specified animal types (golden retriever, tabby kitten, bunny, fox).
- + Beautiful lighting with visible god rays and dew sparkles as requested.
- − The kitten's anatomy where it meets the fox is slightly cluttered and confusing.
- − The butterflies look a bit pasted on/static compared to the dynamic movement of the animals.
Imagen 4.0 Fast Generate 001
- + Clean, high-resolution textures on the fur of each animal.
- + Consistent and warm golden hour lighting.
- − Missed the 'playfully chasing butterflies' and 'tumbling together' action entirely, as the animals are just sitting.
- − Failed certain animal descriptions: the kitten is solid black rather than tabby, and the dog is likely a spaniel or collie mix rather than a golden retriever.
- − Lacks the requested 'dew sparkles' and clear 'god rays' found in the other image.
Verdict: FLUX.2 [dev] Turbo is the clear winner as it followed every detail of the prompt, including the specific breeds/patterns and the complex action of the animals playing together. Imagen 4.0 Fast Generate 001 produced a high-quality but static portrait that failed to include the butterflies, the tabby pattern, or the golden retriever breed.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent typography with perfect spelling and accent on the word 'Caffè'.
- + Beautiful vintage texture and high-quality line work on the banner and cloche.
- + Strong adherence to all prompt elements, including the specific 'Est. 1720' request.
- − The steam elements are a bit thick compared to the fine detail in the rest of the logo.
Imagen 4.0 Fast Generate 001
- + Clean layout that reflects a minimalist vector style.
- + Accurate text rendering for the primary restaurant name.
- + Good use of color palette as requested.
- − Includes hallucinations of extra text ('AFED', 'CARO') above the name.
- − The steam effect is very faint and lacks the defined vector style of the rest of the logo.
- − Banner design is simpler and less elegant than the rival model.
Verdict: FLUX.2 [dev] Turbo followed the prompt with much higher fidelity, producing a professional-grade logo with perfect text and a cohesive vintage texture. Imagen 4.0 Fast Generate 001 produced a decent layout but failed by including unnecessary gibberish text and less refined graphical elements. FLUX.2's execution of the 'Est. 1720' banner and the overall vector emblem style is superior.
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 [dev] Turbo
Imagen 4.0 Fast Generate 001
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Excellent text rendering and spelling for labels like 'Translunar' and 'Descent'.
- + Rich, detailed illustrations of the Saturn V and Lunar Module that remain crisp.
- + Followed the specific color palette instructions well with a deep navy background.
- − Layout is a bit cluttered with redundant 'Translunar' and 'Earth Orbit' icons.
- − Literal interpretation of 'Saturn Vicon' as text instead of recognizing it as 'Saturn V icon'.
Imagen 4.0 Fast Generate 001
- + Very clean, minimalist vector aesthetic that perfectly matches the 'flat-vector' request.
- + Logical flow with a central timeline/trajectory graphic.
- + Excellent adherence to the color palette on a white background.
- − Significant spelling errors throughout including 'APOLO', 'MOOR', and 'VICON'.
- − The icon for 'Launching' shows a moon with a ring instead of a rocket.
- − The sequence flow is confusing and non-linear compared to the prompt's steps.
Verdict: FLUX.2 [dev] Turbo produces a much more informative and legible poster with high-quality illustrations and near-perfect text. While Imagen 4.0 Fast Generate 001 captures the minimalist vector style more accurately, its frequent spelling mistakes and nonsensical iconography make it less useful as an infographic.
Next steps
Explore each model
Google's Imagen 4.0 Fast model optimized for speed and efficiency, suitable for high-volume image generation tasks