AI models
ULM-FiT
Training compute
2.7×10¹⁷ FLOP
Parameters
441M
Published
Jan 18, 2018
ULM-FiT is an AI model developed by University of San Francisco, Insight Centre NUI Galway and Fast.ai (United States and Ireland), first published in January 2018. It works in the language domain, on tasks such as text classification.
Training it took an estimated 2.7×10¹⁷ FLOP of compute (estimation method: operation counting). The model has 441,000,000 parameters. It was trained on roughly 103M datapoints.
Access: Open weights (unrestricted). Its weights are openly available. It is built on top of AWD-LSTM. The reference paper has 1,940 citations. Epoch AI rates the confidence of this record as speculative.
Full record
- Organization
- University of San Francisco, Insight Centre NUI Galway, Fast.ai
- Country of organization
- United States, Ireland
- Domain
- Language
- Task
- Text classification
- Training compute
- 2.7×10¹⁷ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 441,000,000
- Dataset size
- 103M
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Base model
- AWD-LSTM
- Citations
- 1,940
- 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