Papers › Scaling Vision Transformers

Scaling Vision Transformers

8 Jun 2021CVPR 2022 1arXiv:2106.04560archive 2025-07-28

Xiaohua Zhai, Alexander Kolesnikov, Neil Houlsby, Lucas Beyer

Attention-based neural networks such as the Vision Transformer (ViT) have recently attained state-of-the-art results on many computer vision benchmarks. Scale is a primary ingredient in attaining excellent results, therefore, understanding a model's scaling properties is a key to designing future generations effectively. While the laws for scaling Transformer language models have been studied, it is unknown how Vision Transformers scale. To address this, we scale ViT models and data, both up and down, and characterize the relationships between error rate, data, and compute. Along the way, we refine the architecture and training of ViT, reducing memory consumption and increasing accuracy of the resulting models. As a result, we successfully train a ViT model with two billion parameters, which attains a new state-of-the-art on ImageNet of 90.45% top-1 accuracy. The model also performs well for few-shot transfer, for example, reaching 84.86% top-1 accuracy on ImageNet with only 10 examples per class.

Paper PDF Conference PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

google-research/big_vision officialmentioned in papermentioned on GitHubjax report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Few-Shot Image Classification Few-Shot Learning Image Classification

Datasets

Introduced by this paper, per the archive.

JFT-3B

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Image Classification	ImageNet ReaL	ViT-G/14	Accuracy	90.81%	#11 of 57	Archive leaderboard	report
Image Classification	ImageNet V2	ViT-G/14	Top 1 Accuracy	83.33	#6 of 33	Archive leaderboard	report
Image Classification	ObjectNet	ViT-G/14	Top-1 Accuracy	70.53	#14 of 106	Archive leaderboard	report
Image Classification	ObjectNet	NS (Eff.-L2)	Top-1 Accuracy	68.5	#16 of 106	Archive leaderboard	report
Image Classification	VTAB-1k	ViT-G/14	Top-1 Accuracy	78.29	#3 of 34	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position Encodings Adam Attention BPE Dense Connections Dropout Label Smoothing Layer Normalization Linear Layer Multi-Head Attention Position-Wise Feed-Forward Layer Residual Connection Softmax Transformer Vision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections