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 7 shared challenges, with an AI judge weighing in on each.
Nano Banana 2
#1 of 62 in Text-to-Image
Qwen Image Edit 2511
#16 of 32 in Image Editing
Where the votes landed
Nano Banana 2
51.9%
win rate
Ties
3.8%
Qwen Image Edit 2511
44.2%
win rate
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
Man and Car in California
Editing“Make a photo of the man driving the car down the California coastline”
AI Judge Analysis
Nano Banana 2
- + Perfect preservation of the specific Rolls Royce car model from the source.
- + High consistency with the man's hair and facial features.
- + Cinematic composition showing the motion of the car on the coastline.
- − The man's outfit is partially simplified compared to the source.
- − The car appears to be right-hand drive while on a right-hand traffic road.
Qwen Image Edit 2511
- + Excellent preservation of the man's specific plaid jacket, scarf, and hairstyle.
- + The perspective creates an intimate 'in-the-car' feel.
- + High-quality rendering of hands on the steering wheel.
- − Failed to preserve the specific car brand/model (Rolls Royce) from the source image.
- − The interior dashboard design changed significantly from the source.
Verdict: Nano Banana 2 is the better edit because it manages to merge both source images—the specific Rolls Royce and the specific person—into a coherent scene. Qwen Image Edit 2511 does a fantastic job of preserving the man's clothing and likeness, but it completely replaces the car with a generic interior, failing the source preservation aspect of the car image.
Pose & Character Mashup
Editing“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”
AI Judge Analysis
Nano Banana 2
- + Excellent adherence to the request, successfully placing the character from Image 2 in the pose and environment of Image 1.
- + High level of consistency in clothing, scarf, and face details from the character reference.
- + Natural-looking hands and feet that maintain the action of the source image.
- − The scarf's positioning is a bit awkward relative to gravity while leaning.
- − The right leg/foot connection to the Ottoman looks slightly detached.
Qwen Image Edit 2511
- + Successfully preserved the background of Image 2.
- − Failed the primary instruction by simply overlaying the two images rather than performing an edit.
- − Major anatomical and perspective errors caused by the collage-style approach.
- − Did not transfer the character to the new pose.
Verdict: Nano Banana 2 followed the complex instructions perfectly, recreating the man from Image 2 in the dynamic yoga-like pose against the yellow background. In contrast, Qwen Image Edit 2511 failed the instruction entirely, merely pasting the woman from Image 1 in front of the man from Image 2.
Outfit Transfer Challenge
Editing“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”
AI Judge Analysis
Nano Banana 2
- + Successfully preserved the unique vitiligo patterns on the person's face and hands.
- + Captured the specific accessories like the watch, ring, and sunglasses.
- + Maintained the original background and structure of the scene.
Qwen Image Edit 2511
- + Successfully added the full outfit including the boots.
- + Matching the lighting and fabric texture of the coat well.
- − Completely failed to preserve the base person, replacing them with a lookalike of the model from Image 2.
- − Ignored the specific vitiligo features which were core to the original subject.
- − Changed the character's pose significantly from the source.
Verdict: Nano Banana 2 followed the complex instructions much more effectively by keeping the original person's identity and vitiligo patterns intact while compositing the clothing onto their frame. Qwen Image Edit 2511 failed the primary objective of preserving the base person, instead generating a new character that resembles the model from the reference clothing image.
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
Nano Banana 2
- + Excellent source preservation with no changes to facial structure or background.
- + Highly realistic hair texture that matches the age and facial hair of the subject.
- + Natural integration with the existing sideburns and temples.
- − None identified.
Qwen Image Edit 2511
- + Successfully followed the instruction to add a full head of hair.
- − The hair has a noticeably painterly, digital texture that contrasts with the photographic quality of the face.
- − The hairline and curly structure appear somewhat artificial and 'pasted on'.
- − Introduced subtle distortions to the shape of the glasses frame.
Verdict: Nano Banana 2 provided a near-perfect edit, adding hair that looks completely natural and matches the lighting and texture of the original photograph seamlessly. Qwen Image Edit 2511 succeeded in adding hair, but it lacked realistic texture and appeared more like a digital painting, slightly degrading the photographic realism of the source image.
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 2
- + Excellent adherence to all prompt elements including hockey and dogs.
- + High quality text rendering with clever puns like 'W-K9 NEWS'.
- + Strong resemblance to the source subject while maintaining caricature proportions.
- − The 'holding a photo' framing is a slight stylistic departure from a direct digital illustration.
Qwen Image Edit 2511
- + Successfully creates a caricature of the subject with exaggerated features.
- + Clear news anchor setting with a professional suit and microphone.
- − Completely misses the 'hockey' requirement.
- − Nonsensical text in the speech bubble and on the microphone.
- − Incoherent details like a lit candle next to a laptop.
Verdict: Nano Banana 2 is the clear winner for its thorough adherence to the prompt, successfully integrating the hockey stick, puck, and multiple dogs into the news anchor scene. It also features coherent and witty text that enhances the humor. Qwen Image Edit 2511 fails to include the requested hockey theme and struggles with distorted text and illogical background objects.
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 2
- + Perfectly captures the Studio Ghibli art style with watercolor textures and soft line work.
- + Preserves the iconic composition and poses while adding Ghibli-esque environmental details like flowers.
- + Excellent translation of facial expressions into an anime aesthetic.
- − The color of the dress was changed from a vibrant red to a more muted pinkish-peach.
Qwen Image Edit 2511
- + Preserves the original colors and character features more accurately.
- + Successfully softens the lighting and applies a clean digital illustration style.
- − The art style is generic digital manhwa/webtoon rather than the requested Studio Ghibli style.
- − Lacks the 'hand-painted' texture specifically requested in the prompt.
Verdict: Nano Banana 2 is the clear winner as it successfully understands and executes the specific 'Studio Ghibli' aesthetic, including hand-painted watercolor textures and cozy background details. Qwen Image Edit 2511 provides a high-quality clean illustration, but it fails to capture the soulful, painterly essence of the Ghibli style, looking more like a standard filter or digital tracing.
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
Nano Banana 2
- + Strong prompt adherence with energetic windswept hair.
- + Expressive facial change adding to the 'energetic' feel.
- + Authentic leaf rendering with motion blur.
- − Noticeable change to the subject's face compared to the source image.
- − Added some dark artifacts/spots near the leaves by the flowers.
Qwen Image Edit 2511
- + Excellent preservation of the subject's original facial features.
- + Seamless hair extension that matches the wind effect.
- + Clean integration of colorful leaves.
- − The hair flow looks slightly more artificial/symmetrical compared to the source.
- − The energetic feel is slightly lower due to keeping the original posed expression.
Verdict: Nano Banana 2 successfully captured the requested 'energetic and lively feel' by changing the subject's expression and creating realistic motion blur, though it deviated more from the source face. Qwen Image Edit 2511 provided a cleaner edit that perfectly preserved the original person's identity, making it a better choice for a strict source preservation edit.
Explore each model
Alibaba's Qwen image editing model for instruction-based image modifications and transformations