AI models
ProteinStructureTransformer
Training compute
7.6×10²¹ FLOP
Parameters
1.1B
Published
Jan 26, 2024
ProteinStructureTransformer is an AI model developed by Max Planck Institute of Biochemistry (Germany), first published in January 2024. It works in the biology domain, on tasks such as protein or nucleotide language model (plm/nlm).
Training it took an estimated 7.6×10²¹ FLOP of compute (estimation method: hardware). The model has 1,137,000,000 parameters. Training ran on 4 NVIDIA H100 SXM5 80GB for about 10 hours.
It is built on top of ESM2-650M. The reference paper has 21 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Max Planck Institute of Biochemistry
- Country of organization
- Germany
- Domain
- Biology
- Task
- Protein or nucleotide language model (pLM/nLM)
- Training compute
- 7.6×10²¹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 1,137,000,000
- Training hardware
- NVIDIA H100 SXM5 80GB
- Chips used
- 4
- Training time
- 10 h
- Training power draw
- 5.5 kW
- Base model
- ESM2-650M
- Citations
- 21
- 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