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
P-Image Edit
#29 of 32 in Image Editing
Where the votes landed
Nano Banana 2
0%
win rate
Ties
0%
P-Image Edit
0%
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
- + Retains the specific details of the man's hair and clothing (plaid coat and scarf) from the source image accurately.
- + Preserves the exact car model and design with high fidelity.
- + The lighting and motion blur on the wheels create a realistic sense of driving.
- − The car appears to be driving on the wrong side of the road for a standard US coastline drive.
- − The man's scale relative to the car is slightly off, making him look a bit large for the driver's seat.
P-Image Edit
- + Captures the iconic California coastline aesthetic with palm trees and mountains effectively.
- + Correctly places the car on the right-hand side of the road.
- + Good integration of shadows and lighting within the car's interior.
- − The man's facial features and clothing details from the source image are significantly less recognizable than in Model A.
- − The car's front headlights and grille have slight distortions compared to the original source car.
Verdict: Nano Banana 2 is the superior model because it successfully preserves the specific identity of the man and the exact details of his clothing from the source image. While P-Image Edit creates a more geographically accurate California road scene, it loses the character details of the subject, which is critical for an image editing task.
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 character consistency for the face, hair, and clothing.
- + Successfully translates the pose from Image 1 to the male subject.
- + Matches the lighting and background of the source environment perfectly.
- − The pose is slightly less extreme/dynamic than the original reference.
P-Image Edit
- + Successfully combines elements from both images.
- − Failed to follow instructions regarding character hair and clothing details.
- − Severe anatomical artifacts, including a foot where a hand should be and mixed limb textures.
- − Hallucinated long hair and red sleeves not present in the character reference.
Verdict: Nano Banana 2 is the clear winner as it successfully placed the character from Image 2 into the pose of Image 1 while maintaining consistent clothing, facial features, and anatomy. P-Image Edit failed significantly, creating nonsensical anatomy and hallucinating features like long hair and red sleeves that were not requested.
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
- + Excellently preserves the original person's face, skin patterns (vitiligo), and sand textures.
- + Maintains the original image's aspect ratio and photo quality.
- + Highly accurate recreation of the specific scarf and coat textures from Image 2.
- − The hands and watch are modified slightly from Image 2, though they remain coherent.
P-Image Edit
- + Successfully includes shoes, fulfilling the 'including shoes' part of the prompt by zooming out.
- + Maintains the character's facial identity reasonably well.
- − Fails to use the scarf pattern from Image 2, replacing it with a generic one.
- − Significantly alters the background and scale of the person relative to the wooden structure.
- − Adds a gold chain that was not present in either source image.
Verdict: Nano Banana 2 is the clear winner for its superior ability to preserve the subject's exact identity and skin details while perfectly replicating the textures and patterns of the clothing from Image 2. While P-Image Edit manages to include shoes by zooming out, it fails on basic instructions by changing the background and ignoring the specific scarf pattern from the source 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
- + Natural hairline that blends perfectly with the forehead skin
- + Excellent source preservation of the face, glasses, and background
- + Highly realistic hair texture and density that matches the beard
- − None notable
P-Image Edit
- + Interpreted 'full and thick' creatively with a large volume of curly hair
- + Successfully preserved the original background and clothing
- − Hair texture looks slightly plastic or artificial
- − The hairline is very blunt and lacks the transition of natural hair
- − The face has been slightly smoothed or softened compared to the gritty detail of the original
Verdict: Nano Banana 2 provided a near-perfect edit, creating a head of hair that looks completely natural and matches the lighting and texture of the original man's beard. P-Image Edit followed the prompt for 'thick' hair by adding a very large afro-style wig, but the lack of a natural hairline and slight changes to facial texture make it less convincing. Nano Banana 2 is the clear winner for its seamless integration and respect for the source image's realism.
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 caricature style that captures and humorously exaggerates the subject's features.
- + Perfectly integrates all requested elements: hockey stick, multiple dogs, and a news anchor desk.
- + Clear and readable themed text like 'W-K9 NEWS' and 'Anchor's puck drop' adds to the humor.
- − The transition from the caricature drawing to the real hand holding it is a bit jarring but creative.
P-Image Edit
- + High-quality digital illustration style with great vibrant colors.
- + Successfully incorporates the hockey arena background and news anchor setup.
- + Maintains high facial resemblance while still being stylized.
- − The 'dogs' element is very weak, with only one oddly small dog and some strange artifacts on the desk.
- − Gibberish text on the screens in the background.
- − The dog has a strange red heart-shaped structure growing out of its back.
Verdict: Nano Banana 2 is the clear winner because it fully embraces the 'humorous caricature' instruction by creating a cohesive scene where hockey and dogs are central to the news anchor's identity. P-Image Edit creates a high-quality visual, but it fails to effectively integrate the dog element and has significant AI artifacts, such as the nonsensical object on the dog's back and garbled background text.
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
- + Excellently captures the Studio Ghibli art style with watercolor textures and soft line work.
- + Preserves the composition and poses of the original meme perfectly.
- + Enhances the background with charming details like wisteria and potted plants that fit the aesthetic.
- − The facial expressions are slightly softened, losing some of the intensity of the man's gaze.
P-Image Edit
- + Successfully translates the scene into a colorful illustration style.
- + Preserves the iconic expressions and poses of the original characters very well.
- + Dreamy background with colorful clouds matches the prompt's request for nostalgic mood.
- − The art style leans more toward generic modern anime/manhwa than the specific hand-painted Ghibli aesthetic requested.
- − The background replacement is a bit generic compared to the architectural detail in the other model.
Verdict: Nano Banana 2 is the clear winner as it masterfully adapts the source image into a specific Studio Ghibli watercolor style, whereas P-Image Edit creates a more generic digital illustration. Nano Banana 2's attention to textured paper effects and scenic European-style architecture much more closely aligns with the nostalgic, hand-painted aesthetic 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
Nano Banana 2
- + Excellent depiction of wind-blown hair and dog fur.
- + Added leaves look realistic in texture and lighting.
- + Changed facial expression to a joyous laugh, enhancing the 'energetic' feel.
- − The facial expression change, while effective, is a significant departure from the source image's identity.
P-Image Edit
- + Successfully added wind effects to the hair while keeping the face similar to the source.
- + Great preservation of the original facial features.
- − The added leaves look like flat, yellow graphic overlays rather than integrated 3D objects.
- − The wind direction on the hair is inconsistent with the gravity and motion of the leaves.
Verdict: Nano Banana 2 is the superior model as it creates a much more cohesive and realistic scene. While it alters the subject's expression more than P-Image Edit, its integration of the flying leaves and the wind-tossed texture of both the woman's hair and the dog's fur feels natural and dynamic, whereas P-Image Edit's leaves look like a low-quality sticker overlay.
Explore each model
PrunaAI's sub-1-second multi-image editing model supporting up to 5 reference images with state-of-the-art quality