AI models
ALBERT-xxlarge
Training compute
2.4×10²¹ FLOP
Parameters
235M
Published
Feb 9, 2020
ALBERT-xxlarge is an AI model developed by Toyota Technological Institute at Chicago and Google (United States), first published in February 2020. It works in the language domain, on tasks such as language modeling/generation and question answering.
Training it took an estimated 2.4×10²¹ FLOP of compute (estimation method: hardware,third-party estimation,operation counting). The model has 235,000,000 parameters. It was trained on roughly 3.3B datapoints. Training ran on 512 Google TPU v3 for about 32 hours. The compute alone is estimated at $4K in 2023 dollars.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 7,447 citations. Epoch AI rates the confidence of this record as speculative.
Full record
- Organization
- Toyota Technological Institute at Chicago, Google
- Country of organization
- United States
- Domain
- Language
- Task
- Language modeling/generation, Question answering
- Training compute
- 2.4×10²¹ FLOP
- Compute estimation method
- Hardware, Third-party estimation, Operation counting
- Parameters
- 235,000,000
- Dataset size
- 3.3B
- Training hardware
- Google TPU v3
- Chips used
- 512
- Training time
- 32 h
- Training power draw
- 471.2 kW
- Training cost (2023 USD)
- $4K
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 7,447
- Epoch confidence
- Speculative
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