AI models
MGK 8 heads (small)
Training compute
3×10¹⁸ FLOP
Parameters
40M
Published
Oct 16, 2021
MGK 8 heads (small) is an AI model developed by FPT Software AI Center, University of California Los Angeles (UCLA), VinUniversity, Deezer Research, Rice University and University of Texas at Austin (Vietnam, United States and France), first published in October 2021. It works in the language domain, on tasks such as language modeling.
Training it took an estimated 3×10¹⁸ FLOP of compute. The model has 40,000,000 parameters. Training ran on 2 NVIDIA A100.
Access: Unreleased. Its weights are not openly released. The reference paper has 39 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- FPT Software AI Center, University of California Los Angeles (UCLA), VinUniversity, Deezer Research, Rice University, University of Texas at Austin
- Country of organization
- Vietnam, United States, France
- Domain
- Language
- Task
- Language modeling
- Training compute
- 3×10¹⁸ FLOP
- Parameters
- 40,000,000
- Training hardware
- NVIDIA A100
- Chips used
- 2
- Training power draw
- 1.6 kW
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 39
- Epoch confidence
- Confident
More from FPT Software AI Center,University of California Los Angeles (UCLA),VinUniversity,Deezer Research,Rice University,University of Texas at Austin
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