AI models
Transformer + GFM
Training compute
7.7×10¹⁸ FLOP
Parameters
185.2M
Published
Dec 1, 2022
Transformer + GFM is an AI model developed by Nanjing University (China), first published in December 2022. It works in the language domain, on tasks such as language modeling.
Training it took an estimated 7.7×10¹⁸ FLOP of compute (estimation method: operation counting,hardware). The model has 185,200,000 parameters. It was trained on roughly 103M datapoints. Training ran on 8 NVIDIA GeForce RTX 3090.
Access: Unreleased. Its weights are not openly released. It is built on top of FAIRSEQ Adaptive Inputs. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Nanjing University
- Country of organization
- China
- Domain
- Language
- Task
- Language modeling
- Training compute
- 7.7×10¹⁸ FLOP
- Compute estimation method
- Operation counting, Hardware
- Parameters
- 185,200,000
- Dataset size
- 103M
- Training hardware
- NVIDIA GeForce RTX 3090
- Chips used
- 8
- Chip-hours
- 1
- Training power draw
- 5.6 kW
- Model accessibility
- Unreleased
- Open weights
- No
- Base model
- FAIRSEQ Adaptive Inputs
- 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