AI models
ViT-G (model soup)
Training compute
3.4×10²¹ FLOP
Parameters
1.8B
Published
Mar 10, 2022
ViT-G (model soup) is an AI model developed by University of Washington, Columbia University, Google, Meta AI and Tel Aviv University (United States and Israel), first published in March 2022. It works in the vision domain, on tasks such as image classification.
Training it took an estimated 3.4×10²¹ FLOP of compute (estimation method: operation counting). The model has 1,843,000,000 parameters. It was trained on roughly 1.8B datapoints.
Access: Open weights (non-commercial). Its weights are openly available. The reference paper has 1,499 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- University of Washington, Columbia University, Google, Meta AI, Tel Aviv University
- Country of organization
- United States, Israel
- Domain
- Vision
- Task
- Image classification
- Training compute
- 3.4×10²¹ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 1,843,000,000
- Dataset size
- 1.8B
- Model accessibility
- Open weights (non-commercial)
- Open weights
- Yes
- Citations
- 1,499
- Epoch confidence
- Confident
SourceEpoch AI, 'AI Models'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/ai-models. Licensed under CC BY 4.0.
← All ai models