Black Forest Labs' open-weights image generation model with frontier performance, available for non-commercial local deployment
Settled by community votes across 6 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev]
24.3
arena score
#19 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Imagen 3.0 Generate 002
21.0
arena score
#36 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.2 [dev]
0%
win rate
Ties
0%
Imagen 3.0 Generate 002
0%
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 [dev]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [dev]
- + Excellent typography rendering with almost perfect spelling of headings like 'APPETIZERS' and 'MAINS'.
- + Strong use of vibrant accent colors as requested in the prompt.
- + Professional high-quality food photography that feels cohesive.
- − The layout is a bit cramped with text overlapping or being too close to the edge of price columns.
- − Minor spelling error in 'PIZZAU' and 'DIZZA'.
- − The vertical split in the middle makes it look more like a pamphlet than a single-page menu design.
Imagen 3.0 Generate 002
- + Perfectly executed 4x4 grid layout that fits the 'menu on a wall' or 'modern booklet' aesthetic.
- + Consistent photographic style across all food items.
- + Strong modern minimalist vibe with clean spacing.
- − Several spelling errors in headings such as 'APPETIEES' or 'MANS'.
- − Repetitive food imagery where many pizza slices look nearly identical.
- − Text density is a bit low for a functional menu.
Verdict: FLUX.2 [dev] produces much higher quality text and more functional-looking menu sections, while Imagen 3.0 excels at the grid composition and overall aesthetic requested. FLUX.2 is the likely winner because its adherence to the 'professional' and 'bold sans-serif' requirements results in a more usable design despite the simpler layout.
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]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [dev]
- + Excellent font rendering and glowing text effects.
- + Great central composition that focuses on the ingredients.
- + Highly photorealistic textures, especially on the grilled patty and lettuce.
- − The 'exploded' effect is a bit conservative with most ingredients clustered tightly.
Imagen 3.0 Generate 002
- + Strong 'exploded' burger effect with wide separation of components.
- + Vibrant color palette and dynamic spark effects.
- + Good integration of the starburst design element.
- − Text on the starburst contains gibberish characters ('LATIUR%').
- − The tomato slices and bun have a slightly plastic, CGI-like appearance compared to Model A.
- − A stray floating black artifact is visible near the center-right cheese slice.
Verdict: FLUX.2 [dev] is the winner because it successfully rendered all requested text accurately with a high-quality glowing effect and produced a more photorealistic burger. While Imagen 3.0 offered a more dynamic 'exploded' layout, it failed on the text details by including gibberish and had less realistic food textures.
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]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [dev]
- + Excellent text rendering with nearly perfect spelling of the entire prompt.
- + Very realistic chalk texture with smudges and natural handwriting variations.
- + Authentic depth of field and café atmosphere in the background.
- − The word 'with' in the second item is partially obscured by a smudge/artifact.
Imagen 3.0 Generate 002
- + Clean, readable layout with a nice wooden frame.
- + Captures the requested handwriting style consistently across the board.
- − Significant spelling errors and repetitive lines (e.g., 'Risottoo', 'Octpsuin', 'Choouip').
- − The text at the bottom repeats itself redundantly.
- − The handwriting looks a bit too perfect, like a digital brush rather than physical chalk.
Verdict: FLUX.2 [dev] significantly outperforms Imagen 3.0 by accurately rendering almost all of the complex text requested in the prompt with high fidelity and realistic chalk textures. Imagen 3.0 struggles with basic spelling and repeats several lines of text, resulting in a less coherent and professional-looking menu.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.2 [dev]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [dev]
- + Excellent depiction of Earth's atmosphere and moonlight which adds to the cinematic quality.
- + High level of detail on the space suit's textures and hardware.
- + Natural-looking light interactions between the cosmic background and the subjects.
- − The horse's front legs have slightly awkward anatomy/posing near the hooves.
- − The shadow on the clouds below does not match the horse's orientation correctly.
Imagen 3.0 Generate 002
- + Beautiful composition with the nebula centered behind the subjects.
- + Clean, sharp rendering of the horse and the astronaut's gear.
- + Great sense of movement in the horse's mane and tail.
- − The transition between the horse's neck and the astronaut's hand/reins is a bit cluttered.
- − Slightly less 'cinematic' lighting compared to the environmental depth in the first image.
Verdict: Both models failed the negative constraint to put the 'horse on top' of the astronaut, instead interpreting the prompt as a standard knight-style riding pose. FLUX.2 [dev] is the winner because it provides a more cohesive cinematic environment with a planetary backdrop, whereas Imagen 3.0 Generate 002 feels more like a subject placed against a generic space background.
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]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [dev]
- + Excellent text rendering and centered layout.
- + Sophisticated PBR materials with realistic textures for fish and rice.
- + Clean, professional composition that matches the 'diorama' aesthetic perfectly.
- − The flag icon is slightly simple but functional.
Imagen 3.0 Generate 002
- + Successfully captured a clean isometric 3D cartoon style.
- + Includes extra elements like chopsticks and wooden slats for variety.
- + Accurate text and flag icon placement.
- − The text is off-center and squeezed into the corner.
- − The 3D clay-like textures look more like plastic than 'realistic PBR materials'.
- − The rice texture is repetitive and less realistic than the competitor.
Verdict: FLUX.2 [dev] followed the prompt more accurately by centering the text and providing high-quality PBR textures that balance realism with a miniature style. Imagen 3.0 failed to center the text as requested and produced a more generic 'clay' look rather than the sophisticated materials requested.
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]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [dev]
- + Excellent depiction of golden hour lighting with clear god rays
- + High fur detail and anatomical accuracy for all animals
- + Included an extra rabbit which adds to the 'tumbling together' vibe
- − The animals are largely sitting and posing rather than 'playfully chasing' or 'tumbling'
- − The background foliage feels slightly more blurred and less integrated than the foreground
Imagen 3.0 Generate 002
- + Successfully captured the 'tumbling' aspect of the prompt with the bunny on its back
- + Excellent interaction between the animals, creating a more cohesive group dynamic
- + Dew sparkles and particles are very prominent and add to the atmosphere
- − The fox kit has slightly unusual proportions around its paws/body merger
- − The lighting is consistent but lacks the distinct 'god ray' beams requested in the prompt
Verdict: Both models followed the prompt well, but Imagen 3.0 Generate 002 captured the requested 'tumbling' action more effectively, creating a more playful scene. FLUX.2 [dev] produced a more traditional high-quality portrait with superior lighting and god rays, but the animals appear more static and posed.
Next steps
Explore each model
Google's Imagen 3.0 text-to-image generation model, producing high-quality images with improved detail and lighting