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 1 shared challenge, with an AI judge weighing in on each.
Nano Banana 2
#1 of 44 in Text-to-Image
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
P-Image Edit
#24 of 23 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.
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.
Explore each model
PrunaAI's sub-1-second multi-image editing model supporting up to 5 reference images with state-of-the-art quality