Black Forest Labs' open-weights multimodal flow transformer for in-context image generation and editing, available for non-commercial use with character consistency and style transfer capabilities
Settled by community votes across 17 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [dev]
#45 of 48 in Text-to-Image
Nano Banana 2 Lite
#27 of 48 in Text-to-Image
Where the votes landed
FLUX.1 Kontext [dev]
50.0%
win rate
Ties
0.0%
Nano Banana 2 Lite
50.0%
win rate
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.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent adherence to the glass cube requirement with clear edges.
- + Very high-quality textures on the book and sphere, featuring realistic reflections.
- + Accurate lighting that clearly originates from the left window.
- − The plant in the background is less distinct and lacks the 'partially visible through glass' detail requested.
- − The sphere appears to be resting on a mirrored base rather than inside a hollow cube.
Nano Banana 2 Lite
- + Perfectly captures the request for the plant to be visible through the glass.
- + The sphere has a creative floating effect within the solid-looking cube.
- + Excellent text rendering on the book spine that relates to the prompt.
- − The cube looks more like a solid block of acrylic than a glass cube.
- − The lighting on the cube faces is a bit flat compared to the surrounding environment.
Verdict: FLUX.1 Kontext [dev] produced a more photorealistic image with superior textures and lighting, whereas Nano Banana 2 Lite followed the specific layout instructions more accurately, particularly regarding the plant's visibility through the glass. FLUX.1 Kontext [dev] is the likely winner due to its overall visual polish and more realistic representation of the materials.
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
FLUX.1 Kontext [dev]
- + Excellent skin texture and face rendering
- + Strong adherence to the red bicycle prompt requirement
- + Clean, high-resolution visual quality
- − The subject is posing with the bike rather than repairing it as requested
- − Composition is very centered and lacks the 'candid' or 'imperfect framing' feel
- − Missing motion blur from passing cars
Nano Banana 2 Lite
- + Perfectly captures the 'repairing' action with tools and pose
- + Excellent adherence to atmospheric prompts like motion blur, imperfect framing, and reflections
- + Very convincing candid aesthetic with realistic background details
- − Lower resolution and more artifacts compared to Model A
- − Hand and tool anatomy is slightly messy
- − The bicycle has a non-functional physical structure
Verdict: Nano Banana 2 Lite captures the spirit of the prompt much better, successfully executing the 'repairing' action, motion blur, and the candid street photography aesthetic, despite some technical artifacts. FLUX.1 Kontext [dev] produces a higher quality portrait, but the subject is simply sitting on the bike in the middle of a road, failing to follow the core 'repairing' and 'motion blur' instructions.
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
FLUX.1 Kontext [dev]
- + Excellent metallic reflections and lighting
- + Highly detailed engraving on the chestplate
- + Strong cinematic composition
- − Missed the request for braided hair with beads
- − Skin texture appears slightly too smooth/plastic for 'battle-worn'
Nano Banana 2 Lite
- + Perfect adherence to specific details like braided hair with beads
- + Realistic skin texture with convincing dirt and scarring
- + Excellent material variety with chainmail, leather straps, and engraved plate
- − Image has unusual white vertical bars/noise on the sides
- − Proportions of the neck area and gorget are a bit awkward
Verdict: Nano Banana 2 Lite followed the prompt much more closely, specifically including the braided hair with beads and the layered clothing/leather straps that FLUX.1 Kontext [dev] omitted. While FLUX.1 Kontext [dev] produced a cleaner, more cinematic lighting effect, Nano Banana 2 Lite captured the 'battle-worn' aesthetic with much more authenticity despite some technical framing artifacts.
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
FLUX.1 Kontext [dev]
- + Strong minimalist aesthetic with bold typography
- + High-quality, vibrant food photography
- + Clean white background that emphasizes the grid
- − Nonsensical text and poor spelling
- − Layout feels more like a magazine spread than a functional menu
- − Fails to clearly delineate the requested pizza section
Nano Banana 2 Lite
- + Excellent structure with clearly defined sections for appetizers, pizza, and mains
- + Readable, largely accurate English text and pricing
- + Vibrant turquoise accents provide a cohesive professional brand
- − Small photos make details hard to see
- − Slightly cluttered feel compared to a true minimalist style
Verdict: Nano Banana 2 Lite is the clear winner for this task as it follows the structural requirements of a restaurant menu perfectly, including the specific sections requested. While FLUX.1 Kontext [dev] has superior image quality for the food items, it fails as a functional design due to garbled text and a lack of clear menu organization.
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.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Clean, professional graphic design layout
- + Vibrant, high-contrast colors
- + Clear and legible large title text
- − Failed the 'exploded burger' requirement; the burger is fully assembled
- − Spelling error in the secondary text ('LNHLY')
- − The price starburst is a flat graphic rather than a glowing fiery effect
Nano Banana 2 Lite
- + Excellent adherence to the 'exploded burger' requirement
- + Superior fiery/glowing text effects as requested
- + Highly photorealistic textures on food components
- − The text is slightly thin compared to the bold composition
- − Includes extra ingredients like bacon and onion not specifically requested
Verdict: Nano Banana 2 Lite is the clear winner for following the complex structural requirements of the prompt, successfully rendering the 'exploded' view of the burger and the fiery text effects. FLUX.1 Kontext [dev] produced a standard assembled burger and suffered from significant spelling errors in the secondary text.
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.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Successfully renders a chalkboard texture and frame.
- + Handwritten aesthetic is present.
- − Several spelling errors including 'Mashroom', 'Risoktso', and 'Octpus'.
- − Major logic errors in text repetition and garbled date line.
- − The layout is cluttered and lacks the requested elegant cursive header.
Nano Banana 2 Lite
- + Superior text rendering with near-perfect spelling of all menu items.
- + Excellent cafe background composition providing strong environmental context.
- + Beautifully follows formatting instructions including the year, prices, and elegant cursive title.
- − The 'All items made fresh daily' line at the bottom looks more like a digital font than chalk.
- − Slightly less 'natural variation' in letter size compared to hand-drawing, though highly legible.
Verdict: Nano Banana 2 Lite is the clear winner as it followed almost every specific instruction, maintaining perfect spelling and a sophisticated layout. FLUX.1 Kontext [dev] struggled significantly with the text, producing several misspellings and repetitive words that made the menu illegible.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent adherence to the 'horse on top' spatial instruction.
- + Clean, high-fidelity textures on the suit and horse fur.
- + Good anatomical lighting consistency between the two subjects.
- − The composition is a bit static and plain.
- − The horse is more floating behind/over the astronaut rather than explicitly 'riding' him.
Nano Banana 2 Lite
- + Highly cinematic and surreal background with vibrant galaxy details.
- + Creative interpretation with the horse having its own space gear.
- + Dynamic, epic composition.
- − Completely failed the negative instruction; the astronaut is riding a vessel with the horse above, rather than the horse riding the astronaut.
- − Anatomical issues where the astronaut's legs and the vessel blend together.
Verdict: FLUX.1 Kontext [dev] followed the difficult spatial prompt much more accurately, placing the horse on top of the astronaut as requested. Nano Banana 2 Lite produced a more visually stunning and 'cinematic' image, but failed the primary logical constraint of the prompt by effectively having neither riding the other directly and keeping the human in the 'pilot' role.
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.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent fur texture and lighting on the capybara's face.
- + Accurately depicts the yellow taxi driver cap requested in the prompt.
- + Captures a very calm and professional expression on the animal.
- − The capybara only has one paw on the steering wheel, failing the 'both paws' instruction.
- − The human passenger appears to be sitting next to the driver rather than in the back seat.
Nano Banana 2 Lite
- + Perfectly captures the composition with the passenger clearly in the back seat.
- + Follows the instruction for both paws to be placed on the steering wheel.
- + Includes excellent environmental details like the taxi meter and Manhattan skyline.
- − The hat is a black police-style cap instead of the requested yellow taxi cap.
- − The interior of the taxi looks significantly more worn/dirty than a typical professional scene.
Verdict: Nano Banana 2 Lite is the superior choice because it correctly handles the complex spatial relationship between the driver and the back-seat passenger, whereas FLUX.1 Kontext [dev] places them side-by-side. Nano Banana 2 Lite also followed the specific pose instruction for 'both paws' on the wheel, resulting in a more convincing and cohesive scene despite the incorrect hat color.
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.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Successfully included primary and secondary prompt text elements.
- + Clean, legible font for the main title.
- + Contains the requested date and time exactly.
- − Garbled and unreadable text on the scroll banner and location.
- − Simple, flat background that lacks the requested 'moody night sky' and 'parchment' texture.
- − The character illustrations for the bats and pumpkin are overly simplistic.
Nano Banana 2 Lite
- + Excellent adherence to all visual style descriptors including parchment, webs, thorns, and moody sky.
- + Superior text rendering with perfect accuracy on all requested lines including fine print on the banner.
- + Atmospheric cinematic lighting and a highly detailed gothic composition.
- − None noticed; the image matches all prompt requirements perfectly.
Verdict: Nano Banana 2 Lite produced a much more sophisticated and professional invitation, capturing the vintage gothic aesthetic perfectly with clear, accurate text across all fields. FLUX.1 Kontext [dev] struggled significantly with the text on the scroll and location, and the graphics felt more like a basic digital illustration than a cinematic poster.
Bald man challenge
Image 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
FLUX.1 Kontext [dev]
- + Excellent addition of thick, dense hair that follows the head shape well.
- + Preserves the overall lighting and environmental context.
- − Significantly alters the person's facial features and glasses, resulting in a different person.
- − The hair texture looks slightly too smooth or 'AI-stylized' compared to the gritty source image.
Nano Banana 2 Lite
- + Perfect preservation of the original person's facial features, glasses, and jacket details.
- + The hair texture and color match the existing beard and the rugged aesthetic of the image perfectly.
- + The integration of the sideburns into the hairline is very seamless.
- − The volume of the hair is slightly less 'thick' than requested compared to Model A, though more realistic.
Verdict: Nano Banana 2 Lite is the clear winner because it successfully added realistic hair while perfectly preserving the identity of the person in the source image. FLUX.1 Kontext [dev] added good hair but failed the source preservation requirement by essentially generating a new face that only loosely resembles the original.
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
FLUX.1 Kontext [dev]
- + Excellent typography with clean, bold text and a stylized graphic feel
- + Perfectly centered and minimal composition
- + Clean 3D rendering with soft, toy-like textures
- − The flag icon is an abstract shape rather than the Japanese flag
- − The sushi piece is very simplified, looking more like a toy than a food dish with PBR materials
Nano Banana 2 Lite
- + Accurately depicts the Japanese flag as requested
- + Detailed isometric scene with a variety of sushi types
- + Achieves a high-quality miniature diorama aesthetic with realistic lighting
- − The text is smaller and lacks the 'large bold' impact of the competitor
- − Slightly more cluttered than the 'minimal' request
Verdict: Nano Banana 2 Lite is the superior model for this prompt as it accurately rendered the flag and delivered a more detailed 'miniature diorama' scene while maintaining the 45-degree isometric perspective. Although FLUX.1 Kontext [dev] had stronger typography, its failure to produce a recognizable flag and its overly simplistic sushi makes it less effective overall.
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
FLUX.1 Kontext [dev]
- + Successfully translates the subject's face into a comic art style.
- + Includes a dog and a TV to represent the profession.
- + Preserves the denim shirt from the original photo.
- − Fails to include any reference to hockey.
- − The 'caricature' is more of a standard portrait illustration rather than an exaggerated caricature.
- − The text 'JOB' is literal and lacks humor or creativity.
Nano Banana 2 Lite
- + Excellent adherence to all prompts: TV anchor, dogs, and hockey are all integrated.
- + True caricature style with exaggerated features and humorous expressions.
- + Cleverly combines the themes through visual puns like a bone-microphone and a dog playing hockey on the screen.
- − The facial likeness is slightly more generic than Model A, though still recognizable as the subject.
Verdict: Nano Banana 2 Lite is the clear winner as it followed every part of the complex prompt, including the hockey requirement which FLUX.1 Kontext [dev] completely ignored. Nano Banana 2 Lite also delivered a much more creative and humorous 'caricature' interpretation with clever puns and a full scene, whereas FLUX.1 Kontext [dev] provided a relatively static comic-style portrait.
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
FLUX.1 Kontext [dev]
- + Good lighting and bokeh effect in the background.
- + Clean, simple composition with clear focus on the subjects.
- − Failed to include the fox kit and the baby bunny requested in the prompt.
- − Included three animals instead of the four specified, one of which is a duplicate species.
- − Butterflies lack realistic detail and texture.
Nano Banana 2 Lite
- + Perfect prompt adherence, including all four specific animals: retriever, tabby kitten, bunny, and fox kit.
- + Excellent lighting with visible god rays and dew sparkles as requested.
- + Highly detailed fur texture and variety in wildflower types.
- − The fox kit has a slightly awkward pose with its reaching paw.
- − Composition is very crowded with many elements competing for attention.
Verdict: Nano Banana 2 Lite is the clear winner as it successfully rendered all four specific animals requested in the prompt, whereas FLUX.1 Kontext [dev] missed the fox and bunny entirely. Nano Banana 2 Lite also better captured the atmospheric details like god rays and dew sparkles, creating a much richer and more accurate interpretation of the prompt.
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
FLUX.1 Kontext [dev]
- + Successfully translates the image into a clean modern anime style.
- + Preserves the composition and character poses perfectly.
- + Matches the facial expressions well in a stylized format.
- − The style is more generic modern anime than the specific hand-painted Studio Ghibli aesthetic requested.
- − Colors are somewhat flat and lack the dreamy, nostalgic lighting mentioned.
Nano Banana 2 Lite
- + Excellently captures the soft pastel colors and hand-painted texture of Ghibli films.
- + Translates the background into a dreamy, European-inspired street characteristic of Ghibli settings.
- + Maintains an impressive likeness to the original subjects while applying the artistic filter.
- − The blur on the woman in the red dress is slightly distracting compared to the more defined background.
Verdict: Nano Banana 2 Lite is the clear winner as it accurately interprets all stylistic cues of the prompt, specifically the 'hand-painted textures' and 'dreamy backgrounds' typical of Studio Ghibli. FLUX.1 Kontext [dev] provides a high-quality anime conversion, but it feels more like a standard digital illustration rather than the nostalgic, painterly style requested.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent preservation of the subject's face and original lighting
- + Successfully adds wind effect to the hair and dog's tail
- − The flying leaves are very few and look like small, green geometric shapes rather than natural leaves
- − The motion effect is very subtle compared to the 'energetic and lively' request
Nano Banana 2 Lite
- + Strong adherence to the 'energetic' prompt with significant wind and motion effects
- + Adds a many detailed, realistic flying leaves that create depth
- + The dog's fur also show realistic wind effects
- − The face has slightly changed from the original
- − There is a slight distortion/disconnection at the shoulder of the denim jacket due to the motion effect
Verdict: Nano Banana 2 Lite is the clear winner for this edit as it significantly transforms the energy of the image with realistic leaves and strong hair movement. FLUX.1 Kontext [dev] preserved the original subject better but failed to deliver the 'energetic' feel, providing only a few strangely rendered green specks as leaves.
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
FLUX.1 Kontext [dev]
- + Clarity and cleanliness of text
- + Matches the minimalism of the prompt well
- + Excellent vector-style rendering
- − Missed the requested banner element
- − Cloche dome icon looks a bit like a building dome
- − Typography is too modern/bold for a 1720 establishment
Nano Banana 2 Lite
- + Successfully included the banner element
- + Captures the vintage/retro aesthetic perfectly
- + Features the subtle paper texture requested in the background
- − The 'Est.' text is slightly off-center and inconsistent in size
- − Small decorative swirls at the base of the cloche are slightly asymmetrical
Verdict: Nano Banana 2 Lite is the winner because it followed all prompt instructions, specifically including the 'Est. 1720' banner and a clear cloche dome icon. While FLUX.1 Kontext [dev] produced a cleaner high-resolution image, it failed to include the banner and its typography felt too modern for the requested vintage theme.
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
FLUX.1 Kontext [dev]
- + Successfully uses the requested navy and red palette.
- + Captures a playful flat-vector aesthetic.
- − Numerous spelling errors including the main heading 'APOLO'.
- − The icons are abstract and messy, failing to clearly represent the specific mission steps requested.
- − The layout is cluttered and lacks the professional 'infographic' feel requested.
Nano Banana 2 Lite
- + Excellent adherence to the logical order of steps with clear iconography for each.
- + High-quality typography and legible text throughout.
- + Professional and clean layout that perfectly matches the 'modern vector infographic' prompt.
- − The color palette is slightly washed out compared to a traditional bold NASA navy.
- − The landing step is split slightly awkwardly between two positions on the timeline.
Verdict: Nano Banana 2 Lite produced a superior infographic that is both legible and logically structured, accurately depicting the six requested phases of the Apollo 11 mission. FLUX.1 Kontext [dev] struggled significantly with text rendering, spelling errors, and the clarity of the vector icons, resulting in an unpolished final product.
Explore each model
The lightweight, low-cost variant of Nano Banana 2 (Gemini 3.1 Flash Image). Ultra-low-latency image generation and editing at a fixed 1K resolution, designed for high-volume interactive use cases.