Vision Banana | Google DeepMind

🏆

Vision Banana is a SOTA unified model for both image understanding and generation.

🧠

Generative vision pretraining is an effective paradigm for visual understanding.

🔗

Image generation serves as a universal interface for diverse vision tasks.

Prompt: This image is a per-pixel class labeling of the input. The macaron cakes are represented by (255, 255, 0). The round plates are represented by (255, 192, 128). The slice cakes are depicted in (64, 192, 64). The flowers are shown in (128, 0, 64). The tongs are (255, 0, 192).

Prompt: Generate a visualization image of semantic segmentation, using this color mapping: {"cat ears": <255, 165, 0>, "exit sign": <0, 0, 255>, "background": <125, 0, 125>}

Prompt: Conduct per-class semantic segmentation for the given image. The sitting person are represented by (255, 255, 0). The standing and walking people are represented by (255, 192, 128). The ocean is depicted in (64, 192, 64). The street lights are in (128, 0, 64). The sky is in (255, 0, 192). The fence is in (0, 0, 255). The backpack is in (255, 0, 0).

Prompt: Generate an instance segmentation visualization of this image. Each piece of garlic is colored differently.

Prompt: Generate an instance segmentation visualization of the input image. Segment all the price on the price tags, color them differently.

Prompt: Generate an instance segmentation visualization of this image. Each price tag is colored differently.

Prompt: This image shows segmentation masks for the basketballs from the input image. The background is set to #10aa05. Each basketball instance is represented by a solid circular mask, and a different color is used for each mask.

Prompt: This image shows segmentation masks from the given image. The background is black color. The chef's names in both Chinese and English are rendered as cyan color.

Prompt: A segmentation map image. The stretching cat is rendered in green, the cat that is cleaning itself is in cyan.

Prompt: This image shows segmentation masks from the given image. The background is black color. The game control device is represented by a solid yellow.

Prompt: A segmentation map image. The area that corresponds to the man in pink t shirt is rendered solid white; the other man is rendered in green.

Prompt: A segmentation map of the input image. The pig not in the glass is rendered cyan, and the pig in the glass reflection is rendered yellow.

Prompt: Predict the metric depth of this scene as an image. Visualized in the rainbow colormap.

Prompt: Predict the metric depth of this scene as an image. Visualized in the rainbow (black-red-yellow-green-cyan-blue-violet-white) color palette.

Prompt: Predict the metric depth of this scene as an image. Visualized in the rainbow (black-red-yellow-green-cyan-blue-violet-white) color palette.

Prompt: Generate a metric depth map of the provided image.

Prompt: Generate a metric depth map of the input image.

Prompt: Generate a metric depth map of the input image.

Prompt: Generate a metric depth map of the input image.

Loading point cloud…

Prompt: Predict the surface normal of this scene.

Prompt: Generate a surface normal map of the input image.

Prompt: Generate a surface normal map of the input image.

Prompt: Generate a surface normal map of the input image.

Prompt: Generate a surface normal map of the input image.

SegMan-L
(Non Zero-Shot) APE-D OpenSeeD X-Decoder SAM 3 Vision Banana

SAM 3
(Non Zero-Shot) SAM 3
+ Llama 3.2 (ft)
(Non Zero-Shot) Gemini 2.5 APE-D DINO-X OWLv2 Vision Banana
+ Gemini 3.1 Flash-Lite

HyperSeg
+ Phi2
(Non Zero-Shot) X-SAM
+ Phi3
(Non Zero-Shot) HybridGL Kang
+ LLaVA SAM 3
+ Gemini 2.5 Pro Vision Banana

X-SAM
+ Phi3 3.8B
(Non Zero-Shot) LISA-13B-LLAVA1.5(Non Zero-Shot) SegZero RSVP
+ GPT-4o SAM 3
+ Gemini 2.5 Pro Vision Banana
+ Gemini 2.5 Pro

Methods paired with MLLMs for reasoning.

Depth Pro MoGe-2 UniK3D Vision Banana

Vision Banana does not use camera intrinsics in training or inference.

Marigold StableNormal DSINE Lotus-2 Vision Banana

@article{visionbanana2026,
  title={Image Generators are Generalist Vision Learners},
  author={Gabeur, Valentin and Long, Shangbang and Peng, Songyou and Voigtlaender, Paul and Sun, Shuyang and Bao, Yanan and Truong, Karen and Wang, Zhicheng and Zhou, Wenlei and Barron, Jonathan T and Genova, Kyle and Kannen, Nithish and Ben, Sherry and Li, Yandong and Guo, Mandy and Yogin, Suhas and Gu, Yiming and Chen, Huizhong and Wang, Oliver and Xie, Saining and Zhou, Howard and He, Kaiming and Funkhouser, Thomas and Alayrac, Jean-Baptiste and Soricut, Radu},
  journal={arXiv preprint arXiv:2604.20329},
  year={2026}
}