AI models
RFA-GATE-Gaussian-Stateful Big
Training compute
7.1×10¹⁸ FLOP
Parameters
242M
Published
Mar 3, 2021
RFA-GATE-Gaussian-Stateful Big is an AI model developed by University of Washington, DeepMind, Allen Institute for AI, Hebrew University of Jerusalem and The University of Hong Kong (United States, United Kingdom, Israel and Hong Kong), first published in March 2021. It works in the language domain, on tasks such as language modeling/generation and translation.
Training it took an estimated 7.1×10¹⁸ FLOP of compute (estimation method: operation counting). The model has 242,000,000 parameters. It was trained on roughly 103M datapoints. Training ran on 16 Google TPU v3 for about 3.36 hours.
Access: Unreleased. Its weights are not openly released. The reference paper has 430 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- University of Washington, DeepMind, Allen Institute for AI, Hebrew University of Jerusalem, The University of Hong Kong
- Country of organization
- United States, United Kingdom, Israel, Hong Kong
- Domain
- Language
- Task
- Language modeling/generation, Translation
- Training compute
- 7.1×10¹⁸ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 242,000,000
- Dataset size
- 103M
- Training hardware
- Google TPU v3
- Chips used
- 16
- Training time
- 3 h
- Training power draw
- 14.6 kW
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 430
- 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