AI models
TAPE Transformer
Training compute
3×10¹⁹ FLOP
Parameters
38M
Published
Jun 19, 2019
TAPE Transformer is an AI model developed by University of California (UC) Berkeley, Covariant, Google and Chan Zuckerberg Initiative (United States), first published in June 2019. It works in the biology domain, on tasks such as proteins and protein or nucleotide language model (plm/nlm).
Training it took an estimated 3×10¹⁹ FLOP of compute (estimation method: hardware). The model has 38,000,000 parameters. It was trained on roughly 5.2B datapoints. Training ran on 4 NVIDIA V100 for about 168 hours.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 1,004 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- University of California (UC) Berkeley, Covariant, Google, Chan Zuckerberg Initiative
- Country of organization
- United States
- Domain
- Biology
- Task
- Proteins, Protein or nucleotide language model (pLM/nLM)
- Training compute
- 3×10¹⁹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 38,000,000
- Dataset size
- 5.2B
- Training hardware
- NVIDIA V100
- Chips used
- 4
- Training time
- 168 h
- Chip-hours
- 672
- Training power draw
- 2.5 kW
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 1,004
- 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