Meena
Meena is an AI model developed by Google Brain (United States), first published in January 2020. It works in the language domain, on tasks such as text autocompletion and chat. It counts among the frontier models: the systems trained with the most compute of their moment.
Training it took an estimated 1.1×10²³ FLOP of compute (estimation method: hardware,operation counting,third-party estimation). The model has 2,600,000,000 parameters. It was trained on roughly 53.3B datapoints. Training ran on 1,024 Google TPU v3 for about 720 hours. The compute alone is estimated at $215K in 2023 dollars.
Access: Unreleased. Its weights are not openly released. The reference paper has 1,020 citations. Epoch AI rates the confidence of this record as confident.
- Organization
- Google Brain
- Country of organization
- United States
- Domain
- Language
- Task
- Text autocompletion, Chat
- Training compute
- 1.1×10²³ FLOP
- Compute estimation method
- Hardware, Operation counting, Third-party estimation
- Parameters
- 2,600,000,000
- Dataset size
- 53.3B
- Training hardware
- Google TPU v3
- Chips used
- 1,024
- Training time
- 720 h
- Chip-hours
- 737.3K
- Training power draw
- 942.7 kW
- Training cost (2023 USD)
- $215K
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 1,020
- Epoch confidence
- Confident