AI models
SAF R-CNN
Training compute
1.2×10¹⁹ FLOP
Parameters
138M
Published
Oct 28, 2015
SAF R-CNN is an AI model developed by Beijing Institute of Technology, Sun Yat-sen University, Panasonic R&D and National University of Singapore (China and Singapore), first published in October 2015. It works in the vision domain, on tasks such as object detection. It counts among the frontier models: the systems trained with the most compute of their moment.
Training it took an estimated 1.2×10¹⁹ FLOP of compute (estimation method: operation counting). The model has 138,000,000 parameters. It was trained on roughly 350K datapoints. Training ran on 1 NVIDIA GeForce GTX TITAN X.
Access: Unreleased. Its weights are not openly released. It is built on top of VGG16. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Beijing Institute of Technology, Sun Yat-sen University, Panasonic R&D, National University of Singapore
- Country of organization
- China, Singapore
- Domain
- Vision
- Task
- Object detection
- Training compute
- 1.2×10¹⁹ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 138,000,000
- Dataset size
- 350K
- Training hardware
- NVIDIA GeForce GTX TITAN X
- Chips used
- 1
- Training power draw
- 291 W
- Model accessibility
- Unreleased
- Open weights
- No
- Base model
- VGG16
- 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