AI models
DiffDock
Training compute
7.2×10¹⁹ FLOP
Parameters
20.2M
Published
Oct 4, 2022
DiffDock is an AI model developed by Massachusetts Institute of Technology (MIT) (United States), first published in October 2022. It works in the biology domain, on tasks such as proteins.
Training it took an estimated 7.2×10¹⁹ FLOP of compute (estimation method: hardware). The model has 20,240,000 parameters. It was trained on roughly 4.4M datapoints. Training ran on NVIDIA RTX A6000 for about 432 hours.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 731 citations. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Massachusetts Institute of Technology (MIT)
- Country of organization
- United States
- Domain
- Biology
- Task
- Proteins
- Training compute
- 7.2×10¹⁹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 20,240,000
- Dataset size
- 4.4M
- Training hardware
- NVIDIA RTX A6000
- Training time
- 432 h
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 731
- Epoch confidence
- Likely
More from Massachusetts Institute of Technology (MIT)
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