AI models
PolyNet
Training compute
6.4×10¹⁹ FLOP
Parameters
92M
Published
Nov 17, 2016
PolyNet is an AI model developed by Chinese University of Hong Kong (CUHK) (Hong Kong), first published in November 2016. It works in the vision domain, on tasks such as image classification.
Training it took an estimated 6.4×10¹⁹ FLOP of compute (estimation method: comparison with other models,operation counting). The model has 92,000,000 parameters. It was trained on roughly 1.3M datapoints. Training ran on 32 NVIDIA GeForce GTX TITAN X. The compute alone is estimated at $617 in 2023 dollars.
The reference paper has 282 citations. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Chinese University of Hong Kong (CUHK)
- Country of organization
- Hong Kong
- Domain
- Vision
- Task
- Image classification
- Training compute
- 6.4×10¹⁹ FLOP
- Compute estimation method
- Comparison with other models, Operation counting
- Parameters
- 92,000,000
- Dataset size
- 1.3M
- Training hardware
- NVIDIA GeForce GTX TITAN X
- Chips used
- 32
- Training power draw
- 16.8 kW
- Training cost (2023 USD)
- $617
- Numerical format
- FP32
- Citations
- 282
- Epoch confidence
- Likely
More from Chinese University of Hong Kong (CUHK)
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