Black Forest Labs' precision image generation model with maximum control, reliable text rendering, and complete creative control supporting up to 4MP output
Settled by community votes across 6 shared challenges, with an AI judge weighing in on each.
FLUX.2 [flex]
24.8
arena score
#14 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 [flex]
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 [flex]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography rendering with almost perfect spelling
- + Highly professional and realistic food photography
- + Superior layout logic where text and images align with headers
- − Confusing categorization where a pizza image is placed under a 'Mains' text header
- − Section headers are slightly repetitive
Imagen 3.0 Generate 002
- + Stronger adherence to a strict grid layout
- + Clean minimalist aesthetic that feels very modern
- + Good variety of food imagery
- − Text consists largely of illegible gibberish and symbols
- − Sections such as 'Mains' and 'Pizza' are scattered inconsistently across the grid
- − Lack of clear price formatting
Verdict: FLUX.2 [flex] is the clear winner as it produces a functional menu with legible, correctly spelled text and high-quality food photography that fits a professional standard. While Imagen 3.0 Generate 002 creates an attractive minimalist grid, its reliance on garbled text and illogical section placement makes it less usable for a real-world design task.
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 [flex]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with perfect prompt adherence for color and effects
- + High level of photographic realism in food textures like the patty and sauce
- + Strong sense of motion with crumbs and embers scattering naturally
- − The 'exploded' effect is less pronounced than Model B, with several layers still touching
Imagen 3.0 Generate 002
- + Strong 'exploded' view with clear separation of every ingredient
- + Clean and professional composition with balanced layout
- + Accurate starburst element for the price tag
- − Failed to render the fiery, glowing effect on the text as requested
- − Gibberish text included inside the starburst element
- − Lighting on the burger feels more illustrative and flat compared to the realism of Model A
Verdict: FLUX.2 [flex] produced a much more convincing advertisement by perfectly executing the fiery, glowing text effects and maintaining superior photographic realism in the food textures. While Imagen 3.0 Generate 002 did a better job of separating the burger components for the 'exploded' look, it failed the specific text styling instructions and included illegible artifacts in the price starburst.
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 [flex]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [flex]
- + Perfect text accuracy for all requested menu items and numbers.
- + Very realistic chalk texture with smudges and natural handwriting variations.
- + Coherent lighting and soft bokeh background enhance the 'cozy café' atmosphere.
- − The handwriting, while natural-looking, is very consistent, appearing almost like a custom font.
Imagen 3.0 Generate 002
- + Captures a more 'amateur' chalkboard feel with varied line weights and spacing.
- + Good use of colors for the date and title separators.
- − Significant spelling errors and hallucinations in the menu items (e.g., 'Risottta', 'Octpsuin', 'Choouip').
- − Repeated items and fragmented phrases make the menu nonsensical.
- − The framing is very flat and lacks the 'cozy café' environmental context requested.
Verdict: FLUX.2 [flex] is the clear winner as it followed every text instruction perfectly with zero typos and captured the requested atmosphere. Imagen 3.0 Generate 002 struggled significantly with the text rendering, producing several spelling errors and repetitive lines that deviated from the prompt.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.2 [flex]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [flex]
- + Successfully followed the difficult spatial instruction of having the horse on top of the astronaut.
- + High level of cinematic detail with vibrant colors and rich textures in the nebula.
- + Creative interpretation of the surreal prompt with a piggyback-style arrangement.
- − Anatomical issues where the horse's front legs merge awkwardly into the astronaut's shoulders.
Imagen 3.0 Generate 002
- + Clean, high-resolution rendering with logical lighting.
- + Realistic horse anatomy and detailed space suit.
- − Completely failed the negative constraint/spatial instruction for the horse to be on top.
- − Produces a generic image that lacks the 'surreal' quality requested.
Verdict: FLUX.2 [flex] is the clear winner because it correctly interpreted the specific and challenging prompt instruction to place the horse on top of the astronaut. While Imagen 3.0 Generate 002 produced a visually clean image, it ignored the core spatial requirement of the prompt, resulting in a standard 'astronaut riding a horse' cliché.
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 [flex]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [flex]
- + Excellent center-aligned typography with perfect bold weights as requested.
- + Very high-quality PBR materials giving the sushi a realistic yet clay-like texture.
- + Clean, professional composition that perfectly matches the 'diorama base' and 'solid blue background' instructions.
- − The flag icon is slightly disconnected from the text rather than being part of the 'top-center' grouping.
Imagen 3.0 Generate 002
- + Successfully creates an isometric scene with a detailed wooden diorama base.
- + Good variety of sushi types and creative 3D rendering of the 'cartoon' style.
- + Adds extra elements like chopsticks and a wooden mat to fill the scene.
- − Failed to place text at 'top-center', placing it in the top-left corner instead.
- − Texture quality is more flat and plastic-looking compared to the requested 'refined' PBR materials.
- − The text 'SUSHI' is white on a light background, making it harder to read than model_a's bold black text.
Verdict: FLUX.2 [flex] adhered much better to the specific layout instructions, placing the text and flag in the center as requested. FLUX.2 [flex] also achieved a more sophisticated 'refined texture' look that feels more like a high-end 3D render, whereas Imagen 3.0 Generate 002 ignored the text placement constraints and used a more basic cartoon aesthetic.
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 [flex]
Imagen 3.0 Generate 002
AI Judge Analysis
FLUX.2 [flex]
- + Excellent depiction of dynamic movement with the puppy and kitten in mid-stride
- + Clearer and more dramatic god rays and sunrise lighting
- + Better adherence to the 'chasing' part of the prompt with animals looking at the butterflies
- − The fox's anatomy is slightly elongated in its posture
- − The butterflies look a bit more like stickers rather than being integrated into the 3D space
Imagen 3.0 Generate 002
- + Features a charming 'tumbling' composition with the bunny on its back
- + Very soft, realistic fur textures across all four animals
- + Excellent 'dew sparkles' effect on the grass in the foreground
- − The golden retriever looks more like a young dog than a small 'baby' puppy compared to Model A
- − Less dynamic action; the animals appear to be posing rather than chasing
Verdict: Both FLUX.2 [flex] and Imagen 3.0 Generate 002 produced high-quality, adorable images that followed the complex prompt. FLUX.2 [flex] captures the 'chasing' and 'joyful' energy better through motion, while Imagen 3.0 Generate 002 excels in representing the 'tumbling' and 'soft fur' textures in a more static, portrait-like way.
Next steps
Explore each model
Google's Imagen 3.0 text-to-image generation model, producing high-quality images with improved detail and lighting