AI models
EfficientNetV2-XL
Training compute
9.6×10¹⁹ FLOP
Parameters
208M
Published
Jun 23, 2021
EfficientNetV2-XL is an AI model developed by Google and Google Brain (United States), first published in June 2021. It works in the vision domain, on tasks such as image classification and neural architecture search - nas.
Training it took an estimated 9.6×10¹⁹ FLOP of compute (estimation method: hardware). The model has 208,000,000 parameters. It was trained on roughly 14.2M datapoints. Training ran on 16 Google TPU v3 for about 45 hours. The compute alone is estimated at $104 in 2023 dollars.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 4,324 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Google, Google Brain
- Country of organization
- United States
- Domain
- Vision
- Task
- Image classification, Neural Architecture Search - NAS
- Training compute
- 9.6×10¹⁹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 208,000,000
- Dataset size
- 14.2M
- Training hardware
- Google TPU v3
- Chips used
- 16
- Training time
- 45 h
- Training power draw
- 14.6 kW
- Training cost (2023 USD)
- $104
- Numerical format
- FP16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 4,324
- 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