FLUX.2 [klein] 9B Black Forest Labs Imagen 3.0 Generate 002 Google

Settled by community votes across 6 shared challenges, with an AI judge weighing in on each.

FLUX.2 [klein] 9B

20.6 arena score

#13 of 32 in Image Editing

Skill signature · Image Editing

Imagen 3.0 Generate 002

21.0 arena score

#36 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [klein] 9B

win rate

Ties

Imagen 3.0 Generate 002

win rate

Shared challenges 6

Challenge by challenge

The strongest take from each model on every shared challenge, with the AI judge's read.

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 [klein] 9B

Imagen 3.0 Generate 002

AI Judge Analysis

FLUX.2 [klein] 9B

+ Stronger adherence to the 'clean white background' request
+ Header font is bold and high-impact
+ Better overall legibility of the menu list structure

− Confusing repetition of section headers (Appetizers appears twice side-by-side)
− The food variety is low, showing almost exclusively pizza even for the Mains section

Imagen 3.0 Generate 002

+ Excellent variety of food images consistent with a casual dining menu
+ Strict adherence to the 'food photos in grid' instruction
+ Clean and professional 4x4 grid layout maintains high visual interest

− The grid background has a greyish tint instead of the requested pure white background
− Text is largely gibberish/placeholder characters rather than recognizable words

Verdict: Imagen 3.0 Generate 002 is the superior choice for a design mock-up because it captures the 'food photos in grid' requirement perfectly with a professional layout and diverse food items. FLUX.2 [klein] 9B followed the background color instructions better, but suffered from repetitive headings and a lack of variety in the food imagery.

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 [klein] 9B

Imagen 3.0 Generate 002

AI Judge Analysis

FLUX.2 [klein] 9B

+ Excellent typography with perfect adherence to the fiery, glowing effect prompt
+ Features a more dynamic 'exploded' composition with a realistic sense of tilt and motion
+ High-quality photorealistic textures on the beef patty and fresh lettuce

− The burger is not as fully 'separated' into individual layers as is typical for an exploded view advertisement

Imagen 3.0 Generate 002

+ Perfectly separated 'exploded' layers showcasing every individual component clearly
+ Accurate inclusion of all requested text elements, including the price starburst
+ Strong color contrast between the cool lettuce and fiery background

− Includes significant gibberish text inside the price starburst
− The 'Magic Burger' text lacks the high-quality fiery, glowing render compared to Model A
− The shadows on the bottom bun feel slightly disconnected from the ground plane

Verdict: FLUX.2 [klein] 9B produced a more professional and visually appealing advertisement with superior typography and a more cohesive artistic style. While Imagen 3.0 Generate 002 did a better job of separating the burger layers for the 'exploded' effect, it suffered from hallucinated text inside the price badge and less impressive lighting effects on the 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.”

FLUX.2 [klein] 9B

Imagen 3.0 Generate 002

AI Judge Analysis

FLUX.2 [klein] 9B

+ Excellent adherence to the requested chalk handwriting style with a natural, dusty texture.
+ Correctly spelling and formatting the complex menu items requested in the prompt.
+ Effective use of background smudging and chalk dust to enhance the realism of a used chalkboard.

− Slight typo in the bottom row where 'fresh' is spelled 'fress'.
− The lighting at the top of the frame is a bit harsh, obscuring a small portion of the board.

Imagen 3.0 Generate 002

+ Good contrast between the board and the wooden frame.
+ Clear, legible spacing between the lines of text.

− Significant spelling errors and repetitions, such as 'Risottto' and 'Octpsuin & ebbs'.
− Fails to maintain the 'elegant cursive' requested for the title, using a blocky outline style instead.
− The handwriting looks more like a digital font than natural, varying chalk strokes.

Verdict: FLUX.2 [klein] 9B followed the prompt instructions much more accurately, producing realistic, consistent chalk handwriting and correctly spelling the majority of the complex menu items. In contrast, Imagen 3.0 Generate 002 suffered from significant spelling hallucinations and failed to capture the specific 'elegant cursive' style for the header, resulting in a less authentic appearance.

The Reversed Rodeo

Text-to-Image

“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”

FLUX.2 [klein] 9B

Imagen 3.0 Generate 002

AI Judge Analysis

FLUX.2 [klein] 9B

+ Excellent world-building with diverse planetary and nebular details
+ Highly realistic textures on the horse's coat and muscle definition
+ Strong cinematic lighting with reflections on the astronaut's visor

− Failed the subtle 'horse on top' spatial logic request by rendering a standard rider
− Includes distracting gibberish AI-generated text at the bottom
− A few anatomical oddities on the horse's back hooves

Imagen 3.0 Generate 002

+ Clean, focused composition with a striking white horse
+ The nebula background creates a beautiful backlighting effect
+ High clarity and fewer distracting artifacts than its competitor

− Failed the 'horse on top, not vice versa' logic flip requested in the prompt
− Slightly less 'high-detail' depth compared to the environmental density of Model A

Verdict: Both models failed to interpret the specific surreal request for the 'horse on top' (implying the horse riding the astronaut), instead providing traditional interpretations of an astronaut on a horse. FLUX.2 [klein] 9B is the winner due to its superior environmental detail, richer cinematic lighting, and more complex background, even though it included unwanted text.

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 [klein] 9B

Imagen 3.0 Generate 002

AI Judge Analysis

FLUX.2 [klein] 9B

+ Excellent typography rendering and placement
+ Beautiful wood grain texture on the diorama base
+ High-quality soft lighting and shadows

− The flag icon is incorrect (Yemen/Generic tricolor instead of Japan)
− The sushi pieces look slightly disorganized/stacked

Imagen 3.0 Generate 002

+ Correct Japanese flag icon
+ Very clean 3D cartoon aesthetic with consistent textures
+ Excellent layout of various sushi types

− Text layout is left-aligned instead of the requested top-center
− The chopsticks/chopstick rest have some minor geometry intersections

Verdict: Both models followed the isometric 3D styling perfectly. Imagen 3.0 provided the correct flag and a more appealing sushi arrangement, while FLUX.2 produced superior wood textures and followed the text alignment instructions more accurately, despite hallucinating the wrong flag.

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 [klein] 9B

Imagen 3.0 Generate 002

AI Judge Analysis

FLUX.2 [klein] 9B

+ Strong god rays and vibrant backlighting.
+ Excellent fur texture and detail on all animals.
+ Dynamic poses that look like active play.

− Failed to include the baby bunny requested in the prompt.
− The kitten's front paws and anatomical connection to the puppy's body look slightly confused.

Imagen 3.0 Generate 002

+ Successfully included all four requested animals, including the baby bunny.
+ Captured the 'tumbling' aspect of the prompt more effectively.
+ Natural lighting with realistic dew sparkles in the grass.

− The fox kit has a slightly awkward extra-short front limb.
− Overall colors are a bit more muted compared to the '8K masterpiece' aesthetic often expected.

Verdict: While FLUX.2 has slightly better lighting and fur rendering, it missed one of the four key animals requested. Imagen 3.0 successfully included the puppy, kitten, fox, and bunny in a more cohesive composition that matches the 'tumbling together' prompt, despite a minor anatomical glitch on the fox.