AI models
OPT-13B
Training compute
3.5×10²³ FLOP
Parameters
13B
Published
May 2, 2022
OPT-13B is an AI model developed by Meta AI (United States), first published in May 2022.
Training it took an estimated 3.5×10²³ FLOP of compute. The model has 13,000,000,000 parameters. It was trained on roughly 180B datapoints.
Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Meta AI
- Country of organization
- United States
- Training compute
- 3.5×10²³ FLOP
- Parameters
- 13,000,000,000
- Dataset size
- 180B
- Epoch confidence
- Likely
Benchmark results
| 01 | PIQA | Score | 0.76 |
| 02 | HellaSwag | Overall accuracy | 0.7 |
| 03 | BoolQ | Score | 0.65 |
| 04 | WinoGrande | Accuracy | 0.65 |
| 05 | OpenBookQA | Accuracy | 0.4 |
| 06 | ARC AI2 | Challenge score | 0.36 |
| 07 | MMLU | EM | 0.25 |
More from Meta AI
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