AI models
TaLK Convolution
Training compute
2.7×10¹⁹ FLOP
Parameters
240M
Published
Feb 8, 2020
TaLK Convolution is an AI model developed by Carleton University (Canada), first published in February 2020. It works in the language domain, on tasks such as language modeling, translation and text summarization.
Training it took an estimated 2.7×10¹⁹ FLOP of compute (estimation method: operation counting). The model has 240,000,000 parameters. It was trained on roughly 103M datapoints. Training ran on 8 NVIDIA GeForce RTX 2080 Ti 11GB.
Access: Unreleased. Its weights are not openly released. The reference paper has 30 citations. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Carleton University
- Country of organization
- Canada
- Domain
- Language
- Task
- Language modeling, Translation, Text summarization
- Training compute
- 2.7×10¹⁹ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 240,000,000
- Dataset size
- 103M
- Training hardware
- NVIDIA GeForce RTX 2080 Ti 11GB
- Chips used
- 8
- Training power draw
- 4.1 kW
- Numerical format
- FP16
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 30
- 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