AI models
CT-MoS (WT2)
Training compute
5.4×10¹⁷ FLOP
Parameters
45M
Published
Dec 25, 2020
CT-MoS (WT2) is an AI model developed by Google and National Tsing Hua University (United States and Taiwan), first published in December 2020. It works in the language domain, on tasks such as language modeling.
Training it took an estimated 5.4×10¹⁷ FLOP of compute (estimation method: operation counting). The model has 45,000,000 parameters. It was trained on roughly 2M datapoints. Training ran on 4 NVIDIA GeForce GTX 1080 Ti.
Access: Unreleased. Its weights are not openly released. The reference paper has 36 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Google, National Tsing Hua University
- Country of organization
- United States, Taiwan
- Domain
- Language
- Task
- Language modeling
- Training compute
- 5.4×10¹⁷ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 45,000,000
- Dataset size
- 2M
- Training hardware
- NVIDIA GeForce GTX 1080 Ti
- Chips used
- 4
- Training power draw
- 2.0 kW
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 36
- 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