AI models
Subformer (83M)
Training compute
3.5×10¹⁸ FLOP
Parameters
83M
Published
Jan 1, 2021
Subformer (83M) is an AI model developed by University of Tokyo and National Institute of Advanced Industrial Science and Technology (AIST) (Japan), first published in January 2021. It works in the language domain, on tasks such as language modeling/generation and translation.
Training it took an estimated 3.5×10¹⁸ FLOP of compute (estimation method: operation counting). The model has 83,000,000 parameters. Training ran on 8 NVIDIA V100.
Access: Unreleased. Its weights are not openly released. The reference paper has 60 citations. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- University of Tokyo, National Institute of Advanced Industrial Science and Technology (AIST)
- Country of organization
- Japan
- Domain
- Language
- Task
- Language modeling/generation, Translation
- Training compute
- 3.5×10¹⁸ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 83,000,000
- Training hardware
- NVIDIA V100
- Chips used
- 8
- Training power draw
- 4.9 kW
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 60
- Epoch confidence
- Likely
More from University of Tokyo,National Institute of Advanced Industrial Science and Technology (AIST)
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