AI models
Verbatim Memory Transformer (108M)
Training compute
9.9×10¹⁷ FLOP
Parameters
107.7M
Published
Oct 24, 2022
Verbatim Memory Transformer (108M) is an AI model developed by Johns Hopkins University and New York University (NYU) (United States), first published in October 2022. It works in the language domain, on tasks such as language modeling.
Training it took an estimated 9.9×10¹⁷ FLOP of compute. The model has 107,700,000 parameters. It was trained on roughly 40M datapoints. Training ran on 1 NVIDIA Quadro RTX 8000 for about 12 hours.
Access: Unreleased. Its weights are not openly released. The reference paper has 9 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Johns Hopkins University, New York University (NYU)
- Country of organization
- United States
- Domain
- Language
- Task
- Language modeling
- Training compute
- 9.9×10¹⁷ FLOP
- Parameters
- 107,700,000
- Dataset size
- 40M
- Training hardware
- NVIDIA Quadro RTX 8000
- Chips used
- 1
- Training time
- 12 h
- Training power draw
- 286 W
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 9
- Epoch confidence
- Confident
More from Johns Hopkins University,New York University (NYU)
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