AI models
EMDR
Training compute
1.9×10²¹ FLOP
Parameters
440M
Published
Jun 9, 2021
EMDR is an AI model developed by Mila - Quebec AI (originally Montreal Institute for Learning Algorithms), McGill University and DeepMind (Canada and United Kingdom), first published in June 2021. It works in the language domain, on tasks such as question answering.
Training it took an estimated 1.9×10²¹ FLOP of compute (estimation method: hardware,operation counting). The model has 440,000,000 parameters. It was trained on roughly 171.6B datapoints. Training ran on NVIDIA A100 for about 355 hours. The compute alone is estimated at $3K in 2023 dollars.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 191 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Mila - Quebec AI (originally Montreal Institute for Learning Algorithms), McGill University, DeepMind
- Country of organization
- Canada, United Kingdom
- Domain
- Language
- Task
- Question answering
- Training compute
- 1.9×10²¹ FLOP
- Compute estimation method
- Hardware, Operation counting
- Parameters
- 440,000,000
- Dataset size
- 171.6B
- Training hardware
- NVIDIA A100
- Training time
- 355 h
- Training cost (2023 USD)
- $3K
- Numerical format
- FP16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 191
- 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