Nano Banana vs Stable Diffusion 3.5 Large
Head-to-head across 11 challenges
Nano Banana
62.5%
win rate
Ties
6.3%
Stable Diffusion 3.5 Large
31.3%
win rate
Challenge Results
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
Nano Banana
- + Perfect adherence to the spatial requirements of the prompt.
- + Photorealistic lighting and shallow depth of field.
- + Accurate material rendering for the glass, cloth book, and wooden table.
- − The blue sphere appears to be floating inside the cube without support.
- − Slightly less clarity on the 'partially visible' aspect of the plant compared to a real optic glass.
Stable Diffusion 3.5 Large
- + High resolution and sharp details on the glass edges and table grain.
- + Clean rendering of the blue sphere.
- − Failed the spatial prompt; the red book is at the bottom and the sphere is on top of it.
- − The glass cube is around the book and sphere rather than sitting under the book.
- − Lighting is harsh rather than soft.
Verdict: Nano Banana followed every spatial instruction in the prompt, correctly placing the sphere inside the cube and the book on top. Stable Diffusion 3.5 Large failed on the arrangement, placing the book at the bottom and the cube over both the book and sphere.
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
Nano Banana
- + Excellent adherence to technical photography prompts like 50mm lens and motion blur.
- + Realistic skin textures and complex background details like the Japanese lanterns.
- + Logical composition with appropriate tools and props (newspaper, wrench).
- − The bicycle frame geometry is slightly distorted where it meets the rear wheel.
- − The face has a slightly 'smooth' AI quality compared to the background.
Stable Diffusion 3.5 Large
- + Stronger sense of 'candid' street photography with a very natural, unposed feel.
- + Excellent depiction of rain falling and splashing on the ground.
- + Character skin texture feels more organic and age-appropriate.
- − Failed to include 'motion blur from passing cars', as the car in the background is sharp.
- − The bicycle is missing its handlebars, which are replaced by a floating basket and red bars.
- − Anatomy issues with the subject's feet and shoes blending into the pavement.
Verdict: Nano Banana followed the technical aspects of the prompt more closely, successfully incorporating the requested motion blur and shallow depth of field while maintaining a coherent scene. Stable Diffusion 3.5 Large captured a more authentic candid 'vibe' and better rain effects, but suffered from significant anatomical and object errors, specifically the missing handlebars on the bicycle.
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
Nano Banana
- + Excellent adherence to the 'beads' in the hair prompt with clear metallic accents.
- + Superior skin texture with realistic fine wrinkles, pores, and distinct scars.
- + Strong color contrast with vibrant red cloth and clear golden torchlight reflections.
- − The engraving on the armor looks a bit like a flat decal in some areas rather than deep etchings.
- − The background torches look slightly generic and digital.
Stable Diffusion 3.5 Large
- + Very intricate, deep-relief engraving on the plate armor that looks physically carved.
- + The texture of the chainmail/cloth underlayer is exceptionally detailed and lifelike.
- + More naturalistic lighting and background integration for a battlefield setting.
- − The hair braids lack the requested 'small beads'.
- − Eyes appear slightly less 'lifelike' compared to the detail in Model A.
- − The character looks significantly younger and less 'battle-worn' than the description implies.
Verdict: Nano Banana followed the specific details of the prompt more closely, notably including the beads in the hair and a more weathered, battle-worn facial appearance. While Stable Diffusion 3.5 Large featured more impressive technical depth in the armor engravings and fabric textures, it failed to include the requested beads and the character felt slightly too youthful for the requested archetype.
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
Nano Banana
- + Excellent adherence to the grid layout for food photos.
- + Very clean and legible typography with almost correct spelling.
- + Professional colorful borders and accents that enhance the minimalist design.
- − Minor spelling errors like 'APPEITIERS' and 'Brusuechta'.
- − Content mismatch where the description for 'NY Strip' mentions ice cream.
Stable Diffusion 3.5 Large
- + High-quality, appetizing food photography.
- + Artistic composition with large bold primary text.
- − Text is largely unintelligible gibberish throughout the menu sections.
- − The layout feels cluttered compared to the requested modern minimalist aesthetic.
- − The grid photos are cut off at the edges of the frame.
Verdict: Nano Banana followed the prompt much more effectively, producing a usable and clean layout with clear sections and a structured photo grid. While Stable Diffusion 3.5 Large has high visual quality in the food photos themselves, the text is completely illegible and the overall layout fails to meet the professional design requirement.
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
Nano Banana
- + Perfectly adheres to the text placement instructions with 'JAPAN' and 'SUSHI' at the top-center.
- + Clean, minimalist isometric composition that matches the 'soft refined textures' and 'miniature 3D' request.
- + True solid light blue background as specified.
- − The flag icon is a bit large compared to the text and slightly off to the side rather than perfectly integrated.
- − The 3D render style is a bit flat and leans more towards 'cartoon' than 'realistic PBR materials'.
Stable Diffusion 3.5 Large
- + Excellent 3D toy-like aesthetic with vibrant colors and high-quality subsurface scattering effects.
- + Complex miniature details including miniature flags and detailed sushi textures.
- + Effective use of lighting and shadows to create depth on the diorama base.
- − Failed to place text at the 'top-center'; instead, integrated it onto a small physical flag within the scene.
- − The background has a slight gradient shadow, not a perfectly 'solid light blue'.
- − Missing the specific 'JAPAN' then 'SUSHI' below it layout requested for the top of the image.
Verdict: Nano Banana followed the layout instructions much more accurately, placing the text and flag at the top of the frame as requested. While Stable Diffusion 3.5 Large produced a more visually striking 3D miniature with better material depth, it failed the specific framing and text placement requirements of the prompt.
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
Nano Banana
- + Excellent character design with distinct features for all four requested animals.
- + Beautiful lighting with clear god rays and dew sparkles that match the prompt.
- + Central composition captures the playful 'tumbling together' aspect well.
- − The fox's anatomy is slightly strange, with its tail appearing to grow from the side or front.
- − The kitten has an extra paw visible next to its right ear.
Stable Diffusion 3.5 Large
- + Dynamic sense of movement that better captures the 'chasing butterflies' part of the prompt.
- + Natural photorealistic textures on the fur and grass.
- + Stronger 8K masterpiece feel with high-quality bokeh and light interaction.
- − The 'tabby kitten' looks more like a small cougar or caracal kitten, lacking distinct tabby markings.
- − The fox and kitten look very similar in facial structure and color.
Verdict: Both models followed the prompt well, but Nano Banana provided a more accurate representation of the specific animal types requested, particularly the tabby kitten and the fox's distinct tail. However, Stable Diffusion 3.5 Large created a much more dynamic and realistic scene that truly felt like the animals were chasing butterflies in a meadow, whereas Nano Banana looked slightly more posed. Stable Diffusion 3.5 Large is the winner for its superior visual quality and more natural interpretation of the movement.
Victorian Greenhouse Oasis
Text-to-Image“Hyper-photorealistic interior of a lush Victorian glass greenhouse filled with exotic tropical plants, vibrant blooming orchids, tall ferns, colorful butterflies in flight, sunlight filtering through ornate glass roof creating realistic caustics and dew on leaves, intricate iron framework visible, misty atmosphere, 8K masterpiece.”
AI Judge Analysis
Nano Banana
- + Exceptional intricate detail in the Victorian ironwork architecture
- + Dense and varied vegetation that feels lush and overwhelming
- + Superior rendering of textures like mossy cobblestones and varying leaf types
- − The lighting feels slightly flat despite most prompt elements being present
- − Butterflies appear a bit 'stuck on' rather than fully integrated into the lighting of the scene
Stable Diffusion 3.5 Large
- + Beautiful use of volumetric lighting and atmospheric god rays
- + Dynamic composition with a strong focal point at the end of the hall
- + Good scale and size of the orchids in the foreground
- − Architecture feels a bit more generic/gothic rather than ornate Victorian
- − Less variety in the plant species compared to Model A
- − The butterfly in the top right is disproportionately large
Verdict: Nano Banana creates a much more detailed and believable environment with incredible architectural complexity and a vast variety of flora. While Stable Diffusion 3.5 Large has more dramatic lighting and a cleaner 'masterpiece' look, it lacks the sheer density and intricate ironwork requested in the prompt, making Nano Banana the more coherent interpretation of the specific scene.
Heroic Super Hero Portrait
Text-to-Image“Hyper-photorealistic full-body portrait of a female superhero standing triumphantly on a New York skyscraper rooftop at golden sunset, wearing a classic modest superhero costume with flowing cape, chest emblem, gloves, and boots in red and blue colors, practical design, short hair, strong determined heroic expression looking into the distance, powerful confident stance with hands on hips and cape billowing dramatically in the wind, detailed urban cityscape background, warm natural sunlight with sharp shadows and fabric highlights, ultra-sharp textures on suit, hair, and concrete, 8K masterpiece, empowering family-friendly style.”
AI Judge Analysis
Nano Banana
- + Perfect adherence to the 'hands on hips' and 'determined expression' prompts.
- + Excellent anatomical proportions and natural-looking face and hair.
- + Superior depth of field and realistic urban background integration.
- − The cape physics on the left side are slightly disconnected from the shoulder.
Stable Diffusion 3.5 Large
- + High-quality fabric textures on the suit, including metallic sheens.
- + Accurate representation of the golden sunset lighting on the character's face.
- − Failed to follow the 'hands on hips' instruction, instead posing with arms at sides.
- − The perspective of the character standing on the ledge feels slightly flat/greenscreened.
- − The face looks overly airbrushed compared to the rest of the image.
Verdict: Nano Banana followed every detail of the prompt, including the specific pose and the heroic expression, resulting in a more convincing and powerful image. Stable Diffusion 3.5 Large produced a high-quality render with nice material effects but failed the primary pose instruction and had less realistic facial features.
Intricate Floral Mandala
Text-to-Image“Perfectly symmetrical mandala made entirely of real flowers, petals, leaves, fruits, and seeds in vibrant natural colors, intricate layered patterns with radial symmetry, top-down view on a soft neutral background, hyper-detailed organic textures and subtle shadows, photorealistic, 8K masterpiece.”
AI Judge Analysis
Nano Banana
- + Perfect adherence to the 'made entirely of' instruction, using realistic citrus slices, berries, seeds, and ferns.
- + Highly realistic textures and lighting that make the objects look like a physical flat-lay photograph.
- + Excellent radial symmetry across all elements including the small seed patterns.
- − The neutral background texture (linen-like) is slightly repetitive.
- − The overall color palette is a bit muted compared to the 'vibrant' request.
Stable Diffusion 3.5 Large
- + High visual impact with vibrant, punchy colors.
- + Clean, minimalist background that makes the mandala pop.
- + Beautiful central floral illustration with complex layering.
- − The center of the mandala looks like a stylized digital illustration rather than real flowers and petals.
- − Poor radial symmetry; elements like the orange slices and apples are scattered randomly rather than following a symmetrical pattern.
- − Fails the 'made entirely of' prompt by mixing realistic outer fruits with an artificial-looking central graphic.
Verdict: Nano Banana followed the prompt with much higher precision, creating a truly symmetrical mandala using photorealistic organic materials as requested. Stable Diffusion 3.5 Large produced a vibrant and attractive image, but it failed significantly on the symmetry of the outer elements and the central flower lacks the 'real flower' photorealism found in the competitor's output.
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
Nano Banana
- + Perfect text rendering of 'Caffè Florian' and 'Est. 1720'
- + Elegant circular emblem composition following vector logo standards
- + Sophisticated use of negative space for the steam effect
- − The steam is inside/on the cloche rather than rising from it
- − Missing the requested subtle texture on the background
Stable Diffusion 3.5 Large
- + Stronger vintage texture on the light background
- + Dynamic steam effect rising above and below the cloche
- + Rich warm brown and cream tones
- − Spelling error in the main text ('Cafféé' instead of 'Caffè')
- − The cloche is missing its handle on the top and looks slightly disjointed
- − The banner layout is a bit cluttered at the bottom
Verdict: Nano Banana is the clear winner because it successfully renders all requested text accurately and follows a professional vector logo composition. While Stable Diffusion 3.5 Large has better background texture, the spelling error in the primary brand name ('Cafféé') makes it unusable as a logo.
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
Nano Banana
- + Excellent text rendering and spelling accuracy.
- + Follows the sequenced steps and iconography requests perfectly.
- + Clean, modern flat-vector aesthetic with a professional layout.
- − The landing icon is slightly offset from the main horizontal flow.
- − The lunar module illustration is a bit abstract compared to real-world design.
Stable Diffusion 3.5 Large
- + Rich texture and high detail in the lunar surface illustration.
- + Captures the NASA-inspired color palette well.
- − Fails to follow the requested chronological 6-step infographic structure.
- − Text is completely illegible and nonsensical (gibberish).
- − Includes a Space Shuttle-style craft which is historically inaccurate for Apollo 11.
Verdict: Nano Banana followed the prompt with high precision, creating a coherent, readable, and logically structured infographic that correctly identifies the mission steps and crew. Stable Diffusion 3.5 Large produced a visually interesting but chaotic image that failed to follow the requested sequence, used a historically incorrect spacecraft, and featured illegible text.
Nano Banana
Gemini 2.5 Flash Image is optimized for image understanding and generation, offering a balance of price and performance with fast and efficient image generation and editing capabilities.
Stable Diffusion 3.5 Large
Stability AI's 8.1-billion parameter Multimodal Diffusion Transformer (MMDiT) text-to-image model featuring improved image quality, typography, complex prompt understanding, and resource-efficiency