AI models
Cognitron
Training compute
5.2×10⁶ FLOP
Parameters
21.6K
Published
Sep 1, 1975
Cognitron is an AI model developed by Biological Cybernetics (Japan), first published in September 1975. It works in the other domain, on tasks such as miscellaneous image analysis and image classification.
Training it took an estimated 5.2×10⁶ FLOP of compute (estimation method: operation counting). The model has 21,600 parameters. It was trained on roughly 5 datapoints.
The reference paper has 791 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Biological Cybernetics
- Country of organization
- Japan
- Domain
- Other
- Task
- Miscellaneous image analysis, Image classification
- Training compute
- 5.2×10⁶ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 21,600
- Dataset size
- 5
- Citations
- 791
- 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