Gemini 3.1 Flash with image generation capabilities. High-efficiency image generation model with support for text rendering, reference images, search grounding, and thinking mode. The efficient counterpart to Gemini 3 Pro Image.
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
Nano Banana 2
28.4
arena score
#1 of 62 in Text-to-Image
Best Text-to-Image right now
Top 2 in Image Editing
Skill signature
· Text-to-Image
Stable Diffusion 3.5 Large Turbo
10.7
arena score
#61 of 62 in Text-to-Image
Vote tally
Where the votes landed
Nano Banana 2
100.0%
win rate
Ties
0.0%
Stable Diffusion 3.5 Large Turbo
0.0%
win rate
100.0%
0.0% ties
0.0%
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent adherence to the complex spatial prompt.
- + High photorealistic quality with natural textures and light.
- + Accurate rendering of the plant behind and visible through the glass.
- − The sphere is slightly large rather than 'small' as requested.
Stable Diffusion 3.5 Large Turbo
- + Clean, modern aesthetic with sharp lines.
- + Good lighting and shadow work on the table.
- − Failed the spatial request; the book is inside the cube rather than on top.
- − The plant is to the side rather than behind the cube.
- − The image has a more 'rendered' CGI look compared to the requested soft photographic style.
Verdict: Nano Banana 2 followed every spatial instruction perfectly, correctly placing the plant behind the glass and the book on top. Stable Diffusion 3.5 Large Turbo failed notably on the prompt adherence, placing the book inside the cube and the plant to the side.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Expertly captures the atmosphere with realistic reflections on wet pavement and a cinematic 50mm feel.
- + Highly detailed and realistic character features including natural skin texture and appropriate work clothes.
- + Excellent adherence to complex environmental prompts like shop signs, rain, and tools.
- − The wrench in the man's hand is anatomically awkward and doesn't quite connect with the bike chain correctly.
Stable Diffusion 3.5 Large Turbo
- + Follows the basic requirement of an elderly man with a red bicycle.
- − Fails significantly on realism, with smooth, plastic-like textures and a lack of 'candid' feel.
- − Major anatomical issues with the man's hands and the structure of the bicycle frame.
- − Fails to render the background city environment, reflections, or the specific 'no stylization' request.
Verdict: Nano Banana 2 is the clear winner, producing a highly convincing, atmospheric photograph that perfectly captures the 'candid street photo' aesthetic with complex lighting and environmental details. Stable Diffusion 3.5 Large Turbo fails to meet the realism requirements, producing a plastic-looking, distorted image with significant anatomical and structural errors.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent adherence to the battle-worn aesthetic with realistic dirt and faint scarring.
- + Superior texture rendering on the leather straps, cloth underlayers, and engraved metal.
- + Natural inclusion of the torchlight source and bokeh sparks that integrate with the environment.
- − The hand gripping the sword has some anatomical merging where the thumb meets the hilt.
Stable Diffusion 3.5 Large Turbo
- + Clean, high-contrast lighting on the face and armor.
- + Interprets the braided hair request clearly.
- − The 'dirt' and 'scars' look like flat digital paint rather than integrated skin textures.
- − Skin lacks realistic pore detail, appearing overly smooth and plastic-like.
- − Fails to capture the 'beads' in the hair mentioned in the prompt.
Verdict: Nano Banana 2 is the clear winner as it captures the gritty, battle-worn atmosphere requested with incredible detail in the armor's engravings and the texture of the materials. Stable Diffusion 3.5 Large Turbo produces a much more stylized, 'airbrushed' look that misses several key prompt details like the hair beads and realistic skin imperfections.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent typography with nearly perfect, readable text and prices.
- + Logical grid layout that follows modern graphic design principles.
- + High-quality, realistic food photography that fits the casual dining theme.
- − Minor spelling errors in small descriptions like 'somon' and 'metrons'.
- − Alignment of the food photo grid is slightly asymmetrical between sections.
Stable Diffusion 3.5 Large Turbo
- + Vibrant colors and high-contrast food photos.
- + Creative layout using a top-down photographic style.
- − Text is illegible gibberish and contains typos like 'Mians'.
- − Doesn't actually represent a usable menu design, feeling more like accidental collage.
- − The food items have an artificial, plastic-like appearance.
Verdict: Nano Banana 2 produced a professional, functional restaurant menu with impressive text rendering and a realistic layout that perfectly adheres to the prompt. In contrast, Stable Diffusion 3.5 Large Turbo failed to create a legible document, resulting in a cluttered composition with gibberish text and abstract food representations.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
100% wins
0% ties
0% wins
AI Judge Analysis
Nano Banana 2
- + Perfect text rendering according to the prompt specifications.
- + Excellent exploded view with clear separation of all burger ingredients.
- + Dynamic composition with realistic food textures and convincing motion effects.
- − The lighting on the lettuce is slightly oversaturated compared to the background.
Stable Diffusion 3.5 Large Turbo
- + Strong atmospheric lighting and vibrant fire effects.
- + Creative use of smoke and glowing embers in the background.
- − Complete failure to include any of the requested text elements.
- − The burger is mostly intact rather than being in a dynamic exploded view.
- − The image quality has a more illustrative, less photorealistic appearance.
Verdict: Nano Banana 2 followed the prompt's complex instructions perfectly, delivering an exploded burger with all requested text accurately rendered. Stable Diffusion 3.5 Large Turbo failed to include 'MAGIC BURGER' or the pricing and did not provide the exploded view requested.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Perfect text rendering that accurately follows the complex prompt instructions.
- + Highly realistic chalkboard texture with authentic chalk smudges and dust.
- + Consistent and convincing handwritten style that looks organic rather than digital.
- − The background cafe scene is a bit generic and out of focus.
- − The text is centered very precisely, which slightly reduces the 'hand-drawn' randomness.
Stable Diffusion 3.5 Large Turbo
- + Clean and modern composition with pleasant lighting and decor.
- + Creative layout using columns and graphic Dividers.
- − Failed significantly on spelling, including 'ocpus', 'lerbs', and 'choclde'.
- − Text appears like a digital vector font rather than realistic chalk texture.
- − Incorrect date rendering ('apri 31 20-') and missing most of the requested text.
Verdict: Nano Banana 2 is the clear winner as it perfectly captured all the text requested in the prompt with high fidelity and realistic chalk textures. Stable Diffusion 3.5 Large Turbo struggled with spelling, failed to complete the requested items, and produced text that looked like a digital font rather than handwriting.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Perfectly follows the specific instruction to have the horse on top of the astronaut
- + Highly detailed rendering of the space suit and horse fur
- + Vibrant, cinematic color palette and lighting
- − The astronaut's leg is emerging from the saddle area in a way that is anatomically confusing
Stable Diffusion 3.5 Large Turbo
- + Clean, high-contrast composition
- + Sharp rendering of the astronaut suit
- − Failed the negative constraint entirely; the astronaut is on top of the horse
- − The small Earth/Moon in the background has a strange texture and appearance
- − Flat lighting compared to the opponent
Verdict: Nano Banana 2 successfully followed the complex spatial instruction to place the horse on top of the astronaut, creating a truly surreal and cinematic image. Stable Diffusion 3.5 Large Turbo defaulted to a standard interpretation, failing the primary prompt requirement despite its clean execution.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent attention to detail with the inclusion of a 'Captain Cappy' taxi license and realistic fare meter.
- + High photorealism with complex textures and convincing lighting.
- + Perfectly captures the passenger's bored expression while looking at her phone.
- − The cap is black rather than the requested yellow.
Stable Diffusion 3.5 Large Turbo
- + Successfully followed the instruction for a yellow cap.
- + Clear and vibrant bokeh effects in the background lights.
- − The passenger in the back is blurry and not clearly looking at a phone as requested.
- − The perspective is less immersive, appearing more like a side-profile shot through a window rather than a full scene interior.
- − The capybara's hands/paws on the wheel look slightly morphed and less realistic.
Verdict: Nano Banana 2 is the clear winner due to its incredible world-building and adherence to the character dynamics requested; the addition of the 'NYC Hack License' for the capybara is a brilliant detail. While Stable Diffusion 3.5 Large Turbo correctly colored the hat yellow, it failed to clearly depict the businesswoman's actions and lacked the overall sharpness and narrative depth of the first image.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Perfectly followed all text instructions with 100% spelling accuracy and excellent typographic design.
- + Exquisite vintage gothic aesthetic with rich textures, atmospheric lighting, and intricate border details.
- + Composition is professional and balanced, effectively integrating the scroll banner and background elements.
- − The inclusion of skulls was not explicitly requested, though they fit the gothic theme well.
- − The layout is very dense, which might make it difficult to add any additional information later.
Stable Diffusion 3.5 Large Turbo
- + Includes the requested elements like thorns, webs, and twisted trees in a clear way.
- + The center jack-o-lantern has a nice glow effect.
- − Failed to include the majority of the requested text, including the specific event details and the scroll banner.
- − The 'Halloween Party' text is generic and lacks the requested 'elegant gothic' style.
- − The composition is very simple and lacks the cinematic, polished quality requested.
Verdict: Nano Banana 2 is the clear winner as it successfully rendered every piece of requested text with perfect spelling and high-quality typography. While Stable Diffusion 3.5 Large Turbo failed to include the event details or the scroll banner, Nano Banana 2 created a professional-grade invitation with a rich, atmospheric gothic aesthetic that perfectly matches the 'vintage parchment' prompt.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent typography with perfect spelling of both requested words.
- + High-quality realistic PBR textures on the wood grain and food surfaces.
- + Comprehensive variety of sushi types that look professional and appetising.
- − The scene leans more toward realism than the 'cartoon' style requested.
- − Follows a more traditional layout rather than the 'miniature isometric diorama' aesthetic.
Stable Diffusion 3.5 Large Turbo
- + Strongly adheres to the 3D cartoon miniature diorama style.
- + Accurate isometric perspective and clean geometric shapes.
- + Creative integration of the text into the 3D scene environment.
- − Typos in the text with 'SUSHI' rendered as 'SIIHI'.
- − The flag icon is generic and does not represent the Japanese flag requested.
- − The 'JAPAN' text is placed at the top-left rather than top-center.
Verdict: Nano Banana 2 produces a much higher quality image with perfect text rendering and realistic materials, though it misses the specific 'cartoon' vibe. Stable Diffusion 3.5 Large Turbo captures the playful isometric diorama style better but fails on basic text spelling and prompt-specific icons like the Japanese flag. Nano Banana 2 is the better choice for its clarity, professional polish, and text accuracy.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Successfully includes all four requested animals: puppy, kitten, bunny, and fox kit.
- + Excellent use of lighting with clearly defined god rays and golden sunrise atmosphere.
- + Dynamic composition that conveys movement and the specified 'tumbling' and 'chasing' actions.
- − The fox's front right leg has an anatomically awkward bend at the joint.
Stable Diffusion 3.5 Large Turbo
- + Strong foreground bokeh effect creating a sense of depth.
- + Vibrant colors and clean, high-contrast imagery.
- − Failed to include the bunny and the fox kit, showing only two types of animals.
- − Fur textures look somewhat plastic and airbrushed rather than hyper-photorealistic.
- − Characters appear static and posed rather than playful or tumbling.
Verdict: Nano Banana 2 followed the complex prompt requirements much more accurately, including all four requested animal species within a dynamic and beautifully lit meadow. Stable Diffusion 3.5 Large Turbo failed to include the bunny and fox and produced a more generic, static portrait style that lacked the 'masterpiece' photorealism and movement found in Nano Banana 2.
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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent typography including the requested accent on 'Caffè'.
- + Perfect adherence to all prompt elements including the cloche and banner.
- + Clean, professional vector-style execution with a subtle vintage paper texture.
- − The steam lines are a bit simplistic compared to the rest of the illustration.
Stable Diffusion 3.5 Large Turbo
- + Strong high-contrast shading that gives a bold look.
- + Creative interpretation merging a cloche with a coffee mug.
- − Misspelled the primary text as 'Caffeé Florin'.
- − The 'Est. 1720' text is slightly off-center and cluttered within the double banner.
- − Visual style leans more towards a modern illustration than a vintage minimalist logo.
Verdict: Nano Banana 2 followed the prompt perfectly, producing a professional-grade vintage logo with accurate spelling and a cohesive circular composition. Stable Diffusion 3.5 Large Turbo struggled with the spelling of 'Caffè' and produced a more cluttered design that felt less like a minimalist 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.”
Nano Banana 2
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
Nano Banana 2
- + Excellent adherence to the specific iconography and 6-step chronological structure requested.
- + Perfect text rendering for all labels including 'Apollo 11' and crew names.
- + Clean vector aesthetic with a highly consistent color palette as specified.
- − Includes a fourth unknown figure in the crew section that doesn't match the historical three-man mission.
Stable Diffusion 3.5 Large Turbo
- + Features a visually interesting and dynamic layout with a strong retro-poster feel.
- + Good use of the navy and muted red color palette.
- − Failed to follow the requested 6-step structure, omitting most specific steps like 'Lunar Orbit' and 'Descent'.
- − Poor text rendering with several typos and nonsensical filler text.
- − Graphic style is more illustrative than the requested clean flat-vector infographic.
Verdict: Nano Banana 2 followed the technical requirements of the prompt nearly perfectly, delivering a logical and legible 6-step infographic with crisp vector lines. Stable Diffusion 3.5 Large Turbo failed on the specific content instructions, providing gibberish text and missing the required iconographic sequence.
Next steps
Explore each model
Distilled version of SD 3.5 Large that generates high-quality images in just 4 steps, offering faster inference and reduced costs