AI models
2-layer-LSTM+Deep-Gradient-Compression
Training compute
1.3×10¹⁵ FLOP
Parameters
6M
Published
Dec 5, 2017
2-layer-LSTM+Deep-Gradient-Compression is an AI model developed by Tsinghua University, Stanford University and NVIDIA (China and United States), first published in December 2017. It works in the language domain, on tasks such as language modeling.
Training it took an estimated 1.3×10¹⁵ FLOP of compute (estimation method: operation counting). The model has 6,020,000 parameters. It was trained on roughly 929K datapoints.
Access: Unreleased. Its weights are not openly released. The reference paper has 1,627 citations. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Tsinghua University, Stanford University, NVIDIA
- Country of organization
- China, United States
- Domain
- Language
- Task
- Language modeling
- Training compute
- 1.3×10¹⁵ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 6,020,000
- Dataset size
- 929K
- Numerical format
- FP32
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 1,627
- Epoch confidence
- Likely
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