AI models
Hyena-2 153M
Training compute
1.9×10¹⁹ FLOP
Parameters
153M
Published
Feb 21, 2023
Hyena-2 153M is an AI model developed by Stanford University, University of Montreal / Université de Montréal and Mila - Quebec AI (originally Montreal Institute for Learning Algorithms) (United States and Canada), first published in February 2023. It works in the language domain, on tasks such as language modeling/generation.
Training it took an estimated 1.9×10¹⁹ FLOP of compute (estimation method: reported,operation counting). The model has 153,000,000 parameters. It was trained on roughly 15B datapoints. Training ran on 8 NVIDIA A100 SXM4 80 GB.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 497 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Stanford University, University of Montreal / Université de Montréal, Mila - Quebec AI (originally Montreal Institute for Learning Algorithms)
- Country of organization
- United States, Canada
- Domain
- Language
- Task
- Language modeling/generation
- Training compute
- 1.9×10¹⁹ FLOP
- Compute estimation method
- Reported, Operation counting
- Parameters
- 153,000,000
- Dataset size
- 15B
- Training hardware
- NVIDIA A100 SXM4 80 GB
- Chips used
- 8
- Training power draw
- 6.4 kW
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 497
- Epoch confidence
- Confident
More from Stanford University,University of Montreal / Université de Montréal,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms)
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