Nano Banana vs Grok Imagine Image
Head-to-head across 18 challenges
Nano Banana
58.1%
win rate
Ties
6.5%
Grok Imagine Image
35.5%
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
- + Excellent lighting and atmospheric dust motes create a realistic photographic feel.
- + The plant's leaves are clearly visible through the glass cube with realistic refraction.
- + Sharp textures on the book cover and the wooden table surface.
- − The blue sphere appears to be floating in the center of the cube rather than resting on a surface within it.
- − The glass cube lacks a clear top face where it meets the book.
Grok Imagine Image
- + Very high clarity on the glass material and its reflections.
- + The composition feels grounded with realistic shadows and textures.
- + Accurately captures the 'partially visible through glass' requirement for the plant.
- − The blue sphere is levitating in the middle of the glass, which looks physically impossible.
- − The cube is more of a rectangular prism than a perfect cube.
Verdict: Nano Banana produces a more artistically pleasing image with beautiful soft window lighting and realistic plant detail seen through the glass. Grok Imagine Image has higher structural clarity in the glass reflections, but the lighting is flatter compared to Nano Banana. Both models failed to place the sphere realistically on the bottom of the cube, opting instead for a levitating appearance.
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 depth of field and lens characteristics.
- + High level of detail in skin texture and environment.
- + Atmospheric lighting with clear rain and reflection effects.
- − The bicycle anatomy is slightly warped with a redundant frame bar.
- − The lighting feels a bit too 'staged' for a candid street photo.
Grok Imagine Image
- + Successfully captures the 'imperfect framing' and 'candid' feel of the prompt.
- + Strong execution of motion blur from passing cars.
- + Highly realistic, unpolished aesthetic that looks like a real snapshot.
- − The subject's face is obscured, making it harder to judge the 'elderly' and 'Japanese' descriptors.
- − Shadows on the subject are a bit muddy and lack detail.
Verdict: Nano Banana produces a beautiful, cinematic image with great detail, but it feels more like a movie still than a candid street photo. Grok Imagine better captures the requested 'candid' and 'imperfect' street photography vibe, including the motion blur of cars, but sacrifices some clarity on the subject's face.
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 photo layout with a colorful, uniform grid that matches the prompt exactly.
- + Text is largely legible with clean, modern sans-serif fonts.
- + High visual quality in food photography with vibrant colors.
- − Several typos in section headers (Appeitiers) and item names (Brusuechta, Margheitte).
- − Dish descriptions are often nonsensical or repetitive (e.g., NY Strip with ice cream).
Grok Imagine Image
- + Layout feels more authentic to a real restaurant menu with multiple columns.
- + Clever integration of food photography interspersed with text.
- + Good use of bold typography and clear section headers.
- − Prompt specified a 'grid' of photos, which was not followed.
- − Text contains significant 'AI gibberish' in the smaller descriptions.
- − Repeats item names multiple times to fill space (Steak Frites/Grilled Salmon).
Verdict: Nano Banana followed the specific layout request for a colorful photo grid much more closely and produced much more legible text, despite several spelling errors. Grok Imagine Image created a more professional-looking traditional layout, but it ignored the 'grid' instruction and the descriptive text is mostly unreadable symbols.
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 depiction of battle-worn textures including rust, pits, and scars.
- + Rich detail on the intricate eagle engraving and leather strap textures.
- + Strong adherence to the 'beads' in hair prompt with metallic hair rings.
- − The scars look a bit like surface scratches rather than deep healed wounds.
- − The background flame effect is slightly generic.
Grok Imagine Image
- + Stunning cinematic lighting with high contrast and realistic torchlight bloom.
- + The 'small beads' in the hair are very distinct and follow the prompt well.
- + The engraved plate armor is exceptionally intricate and filigreed.
- − The 'dirty' skin looks more like modern makeup smudges than battle grime.
- − The armor lacks the 'worn' and aged look requested, appearing brand new despite the character's scars.
Verdict: Nano Banana excels at capturing the 'battle-worn' aspect of the prompt, providing grit, rust, and skin texture that feels lived-in. However, Grok Imagine Image produces a more aesthetically striking portrait with beautiful lighting, more delicate engraving, and a clearer interpretation of the beads in the hair, even if the armor looks a bit too pristine.
Man and Car in California
Editing“Make a photo of the man driving the car down the California coastline”
AI Judge Analysis
Nano Banana
- + Expertly preserves the car structure from the source image, including the exact wheel and headlight design.
- + Successfully integrates the specific man from the source image, including his dreadlocks and smile.
- + The motion blur on the road and wheels creates a convincing sense of driving.
- − The man is slightly small and obscured by the windshield frame, but still identifiable.
Grok Imagine Image
- + High visual quality and a beautiful scenic representation of the California coastline.
- + Good use of lighting and composition for a commercial-style car photo.
- − Fails to use the man from the source image, replacing him with a standard middle-aged white man.
- − Modifies the car's headlights and body lines significantly compared to the source image.
- − Poor source preservation, essentially generating a new image rather than editing the provided ones.
Verdict: Nano Banana followed the instructions perfectly, taking the specific car and the specific man from the source images and combining them into the requested scene. Grok Imagine Image failed as an editor, ignoring the identity of the man in the source image and altering the car's design, resulting in a generic stock photo rather than a specific edit.
Bald man challenge
Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
Nano Banana
- + Excellent source preservation of original facial features and clothing.
- + Natural texture and density that matches the beard style.
- + Correct integration of lighting on the added hair strands.
- − The hairline on the forehead is a bit high and chaotic.
- − Slightly messy/unkempt style compared to the prompt's request for a full head of hair.
Grok Imagine Image
- + Very realistic and clean hairline integration.
- + Excellent preservation of the background and original clothing.
- + High-quality texture that looks like a natural growth pattern.
- − Slight alteration to the bridge of the nose and eyeglass frames where the hair meets the temple.
Verdict: Both models performed exceptionally well at this inpainting task, preserving nearly all aspects of the source image while adding realistic hair. Grok Imagine Image is the winner because the hairline and overall hair grooming feel more natural and 'full' as requested, whereas Nano Banana's hair looks slightly more like a wig placed on top due to the messy forehead transition.
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
- + Excellent adherence to the '45° top-down isometric' and 'diorama base' requirements.
- + Superior rendering of soft refined textures and miniature cartoon aesthetic.
- + Precise text execution with the flag icon integrated exactly as requested.
- − The text is slightly off-center to the right compared to the plate.
Grok Imagine Image
- + Bright, clean visual style with high clarity.
- + Good adherence to the color palette and text requirements.
- + Varied selection of sushi types within the scene.
- − Failed the 'isometric' request, providing a standard perspective view.
- − Lighting is a bit harsh with strong shadows, missing the 'gentle lighting' prompt.
- − The text spacing and font choice feel slightly more generic than Model A.
Verdict: Nano Banana followed the prompt's technical requirements much better, specifically the isometric perspective and the diorama base, resulting in a cohesive miniature look. Grok Imagine Image produced a high-quality graphic but missed the isometric orientation, opting for a standard perspective that lacked the 3D diorama feel requested.
Night Sky Transformation
Editing“Change the scene to night: a deep, dark sky with subtle, glistening stars visible behind the mountain.”
AI Judge Analysis
Nano Banana
- + Perfect preservation of the original town and foreground details.
- + Subtle, realistic star density that feels natural.
- + Deep, dark sky that accurately reflects a night scene.
- − The transitions around some distant mountain ridges are slightly soft.
Grok Imagine Image
- + Excellent sky color and star visibility.
- + Maintains the structural integrity of the town and landscape perfectly.
- − The stars are a bit more uniform and 'digital' compared to Model A.
Verdict: Both models performed exceptional edits, successfully transforming the sunset sky into a night sky while perfectly preserving the complex village and mountain terrain from the source image. Nano Banana is slightly preferred for its more naturalistic, varied star field, whereas Grok Imagine's stars feel a bit more like a generic overlay, though both are technically very high quality.
Over-the-top cartoon caricature
Editing“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”
AI Judge Analysis
Nano Banana
- + Excellent humor and creative integration of all elements like the dog in a helmet.
- + Distinctive caricature art style with highly exaggerated features.
- + Great text rendering on the microphone and breaking news graphic.
- − The facial likeness is somewhat lost in the heavy caricature style.
- − The muscular arm is a jarring addition that wasn't specifically requested.
Grok Imagine Image
- + Maintains a much stronger facial resemblance to the source image.
- + Very clean, professional digital illustration style.
- + Clear and logical layout of the TV news desk and hockey background.
- − The dog's paw and the woman's hand on the desk have anatomical errors.
- − Less 'exaggerated' and humorous compared to the other model's interpretation.
Verdict: Nano Banana leans heavily into the 'caricature' aspect with great humor and creative details, though it loses some of the subject's likeness. Grok Imagine Image preserves the subject's face much better and provides a cleaner aesthetic, but suffers from significant AI artifacts in the hands and paws.
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
- + Successfully includes all four requested animals accurately.
- + Features clear, detailed butterflies chasing the animals.
- + High level of fur detail and realistic golden hour lighting with dew sparkles.
- − The fox kit's anatomy is slightly awkward where it meets the kitten.
- − The lighting looks somewhat artificial/digitally overlaid in the center sunburst.
Grok Imagine Image
- + Intense, high-contrast lighting creates a dramatic wholesome atmosphere.
- + Excellent texture on the puppy and fox fur.
- − The 'butterflies' appear more like small white moths or bugs.
- − The kitten has bright blue eyes and patterns that look more like a wild cat than a standard tabby.
- − The bunny looks somewhat like a hybrid or a squirrel due to the fur color and facial shape.
Verdict: Nano Banana is the winner because it strictly follows the prompt, accurately portraying all four specific animals and featuring clear butterflies. Grok Imagine Image has beautiful lighting but fails on the details, with the butterflies looking like generic insects and the animals (specifically the kitten and bunny) looking less like their requested breeds.
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 level of detail in the ornate Victorian ironwork and glass roof textures.
- + Highly realistic distribution of a wide variety of plants and orchids.
- + Excellent atmosphere with soft mist and convincing lighting that creates a sense of scale.
- − Butterflies are very small and somewhat blend into the background detail.
- − The overall image has a slightly muted, desaturated color palette compared to Model B.
Grok Imagine Image
- + Vibrant, high-contrast colors make the foliage and butterflies pop.
- + Clear, large butterfly subjects that directly address that part of the prompt.
- + Strong use of light rays (volumetric lighting) filtering through the glass.
- − Lacks the 'hyper-photorealistic' quality requested, appearing more like a digital illustration.
- − The scale of the butterflies is unnaturally large relative to the plants.
- − Ironwork transitions and foliage edges exhibit some digital smoothing and lack fine texture.
Verdict: Nano Banana is the superior choice for this prompt as it captures the specific '8K masterpiece' and 'hyper-photorealistic' requirements with intricate detail in the Victorian architecture and mossy ground. Grok Imagine Image produces a more stylized, illustrative result with unrealistic proportions for the butterflies, though it features more dramatic lighting. Nano Banana's atmospheric rendering and realistic botanical variety much more closely align with the requested Victorian greenhouse aesthetic.
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
- + Excellent adherence to the 'hands on hips' prompt.
- + High-quality skin texture and realistic facial features.
- + Captures the iconic New York City skyline with recognizable landmarks like the Empire State Building.
- − The cape physics look a bit frozen and less dynamic than Model B.
Grok Imagine Image
- + Very dramatic and dynamic cape movement.
- + Strong lighting with a clear sunset glow and backlighting.
- − Failed the 'hands on hips' instruction, showing one hand as a fist.
- − The chest emblem is a distorted version of the Superman logo rather than an original design.
- − The skyline is generic and less detailed compared to Model A.
Verdict: Nano Banana followed the prompt instructions much more accurately, specifically capturing the requested pose and the New York City background. While Grok Imagine Image had more dramatic lighting, it failed on the specific hand placement and used a derivative superhero logo rather than a practical original design.
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
- + Excellent adherence to the 'fruits' and 'seeds' part of the prompt with clear orange slices and chia seeds.
- + Very strong radial symmetry across all layers of the composition.
- + Highly photorealistic textures that make the items look like they were physically laid out on a linen surface.
- − The lighting is somewhat flat across the entire image, lacking deep value contrast.
- − The colors, while natural, are slightly more muted than the 'vibrant' request.
Grok Imagine Image
- + Striking vibrant colors and high-contrast lighting that creates a strong 3D effect.
- + Intricate layered pattern that creates a beautiful, complex focal point.
- + Good variety of botanical elements including petals and textured seeds.
- − Symmetry is imperfect, particularly on the outer edges and secondary floral elements.
- − The composition feels slightly more digital and less like a photograph of real physical objects.
- − The 'fruits' requested are less identifiable compared to Model A's clear citrus slices.
Verdict: Nano Banana wins because it achieved near-perfect radial symmetry and looks authentically like a physical creation made from real botanical items, specifically capturing the fruits and seeds mentioned in the prompt. While Grok Imagine Image has more vibrant colors and a more dynamic 3D feel, it suffers from several asymmetrical elements that detract from the 'perfectly symmetrical mandala' requirement.
Neutral Expression to Genuine Smile
Editing{
"action": "image_edit",
"reference": "uploaded neutral portrait",
"change": "Warm genuine Duchenne smile: lips curved up, slight natural teeth, soft eye crinkles, subtle cheek raise",
"details": "Realistic smiling skin (dimples if present, soft cheek shadows), slightly brighter eyes; keep exact eye shape/color/iris",
"preserve_exact": "Face identity/structure, eyes/nose/lips/eyebrows, hair, skin texture/pores/freckles, makeup, clothing, head pose, background, lighting, shadows, framing",
"no_changes": "No face shape change, no new features, no gaze shift, no hair/clothing/lighting/background edits",
"style": "Ultra-photorealistic 8K portrait, sharp face focus, natural soft lighting, realistic skin glow"
}
AI Judge Analysis
Nano Banana
- + Perfect preservation of hair strands and fine skin textures from the original.
- + Delivers a natural Duchenne smile with convincing eye crinkles and cheek elevation.
- + Maintains the exact eye shape and iris detail while adding a subtle brightness.
- − The transition of the teeth to the lips is slightly sharper than Model B's.
Grok Imagine Image
- + Successfully applies the smile effect while keeping the head pose and background intact.
- + Good light/shadow balance on the moved cheek muscles.
- − Noticeable loss of high-frequency detail in the hair, which appears smoother and more 'painted' than the original.
- − Smudging of skin texture and freckles compared to the original image.
- − The smile looks a bit more forced or 'morphed' rather than organically integrated into the existing face structure.
Verdict: Nano Banana is the clear winner because it achieves the requested edit while perfectly preserving the intricate details of the source image, such as individual hair strands and skin pores. In contrast, Grok Imagine Image applies the smile but introduces unwanted smoothing and artifacts to the hair and skin, losing the high-fidelity realism of the original portrait.
Studio Ghibli Anime Style
Editing“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”
AI Judge Analysis
Nano Banana
- + Excellent preservation of the original image's structure and details
- + Maintains the exact facial expressions and poses of the source
- − Fails to apply the requested artistic style change
- − Only applies a slight softening/blurring effect rather than an illustration style
Grok Imagine Image
- + Strong adherence to the Studio Ghibli illustration style
- + Excellent use of hand-painted textures and soft pastel colors
- + Perfectly captures the dreamy background and nostalgic mood requested
- + Preserves the composition and recognizable features of the source image
- − Slight loss of the specific 'indignant' expression on the woman on the right
Verdict: Nano Banana failed the prompt by only making minor photographic adjustments and ignoring the request for a Ghibli-style illustration. Grok Imagine successfully transformed the image into a beautiful hand-painted anime style while maintaining the iconic composition and characters of the original meme.
Golden Hour Stroll
Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
Nano Banana
- + Excellent hair physics that realistically flow to one side.
- + The added leaves look integrated into the scene's lighting.
- + Preserves the original facial features and dog's appearance almost perfectly.
- − The dog's tail is slightly altered compared to the source image.
- − Motion blur on the leaves is minimal, making them look a bit static.
Grok Imagine Image
- + Vibrant leaf colors and higher density of elements added.
- + Hair is successfully edited to blow in the wind.
- + Successfully preserved the background and overall composition.
- − The hair ends look slightly jagged or digitally 'cut out' in some sections.
- − Leaves appearing in front of the subject's clothing look flat and lack shadows.
- − The dog's left ear has been changed into an unusual shape.
Verdict: Both models followed the instructions well, but Nano Banana (Model A) provides a much more natural-looking edit with better hair physics and subject preservation. Grok Imagine Image (Model B) adds more leaves, but they look like a flat overlay, and it introduces minor artifacts/deformations to the dog's ear.
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
- + Perfectly adheres to all prompt elements including the circular emblem style.
- + Integrated the 'Est. 1720' text into a banner as specifically requested.
- + The vector art is clean, symmetrical, and fits the vintage minimalist aesthetic perfectly.
- − The steam is depicted inside the cloche rather than rising from it, which is a bit literal.
- − The texture on the background is very subtle, almost unnoticeable.
Grok Imagine Image
- + Strong, bold typography that is very legible.
- + Nice use of highlights and shadows to give the cloche dimension.
- + Accurately depicts steam rising from the top of the dome.
- − Failed to include the requested banner for the 'Est. 1720' text.
- − Repeated the 'Est. 1720' text twice, cluttering the composition.
- − Includes odd extra elements like a handle/spoon shape on the right that wasn't requested.
Verdict: Nano Banana followed the prompt much more accurately, successfully incorporating the banner and circular emblem style requested. While Grok Imagine has nice lighting on the cloche, it failed on several specific instructions by omitting the banner, adding unexplained shapes, and repeating the date text twice.
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 typography with almost perfect spelling and readable text.
- + Highly consistent iconography and clean vector style.
- + Perfect adherence to the requested NASA-inspired color palette.
- − Step 5 and 6 use nearly identical icons for the moon.
- − The composition has a lot of empty space in the upper right.
Grok Imagine Image
- + More dynamic composition that fills the frame more effectively.
- + Good interpretation of the descent and landing icons on the moon surface.
- + Correctly followed the numerical steps requested in the prompt.
- − Several significant spelling errors in the labels (e.g., '3rajoory', 'Transluiory', 'Moom').
- − Included the words 'NASA inspired' literally from the prompt into the artwork.
- − Icon for Earth orbit has a Saturn-like ring that feels inaccurate.
Verdict: Nano Banana is the clear winner because it produces a professional, usable infographic with clean typography and consistent iconography that adheres perfectly to the style prompt. While Grok Imagine has a more engaging layout, its text is riddled with typos and it fails the 'crisp lines' requirement by including messy, distorted labels.
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.
Grok Imagine Image
An image generation model by xAI designed to generate highly aesthetic images from text descriptions.