AI models
Neocognitron
Training compute
2.7×10⁸ FLOP
Parameters
1.1M
Published
Apr 1, 1980
Neocognitron is an AI model developed by NHK Broadcasting Science Research Laboratories (Japan), first published in April 1980. It works in the vision domain, on tasks such as character recognition (ocr). It counts among the frontier models: the systems trained with the most compute of their moment.
Training it took an estimated 2.7×10⁸ FLOP of compute (estimation method: operation counting). The model has 1,140,576 parameters. It was trained on roughly 5 datapoints.
The reference paper has 5,782 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- NHK Broadcasting Science Research Laboratories
- Country of organization
- Japan
- Domain
- Vision
- Task
- Character recognition (OCR)
- Training compute
- 2.7×10⁸ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 1,140,576
- Dataset size
- 5
- Citations
- 5,782
- Epoch confidence
- Confident
More from NHK Broadcasting Science Research Laboratories
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