Training compute growth
Maximum disclosed training compute per year on a log10(FLOP) scale. Missing values remain absent; the current year is incomplete.
Training compute growth
Total 468 · Page 11/19 · Observed 9/25/2026, 3:30:45 AM
| Name | Group | Reported value | Date | Notes | Source |
|---|---|---|---|---|---|
| Gato | DeepMind | 4,020,000,000,000,000,000,000 FLOP | 2022-05-12 | Unreleased | Source ↗ |
| wave2vec 2.0 LARGE | 3,870,720,000,000,000,000,000 FLOP | 2020-10-22 | Open weights (unrestricted) | Source ↗ | |
| ViT-G (model soup) | University of Washington,Columbia University,Google,Meta AI,Tel Aviv University | 3,400,000,000,000,000,000,000 FLOP | 2022-03-10 | Open weights (non-commercial) | Source ↗ |
| ELECTRA | Stanford University,Google,Google Brain | 3,100,000,000,000,000,000,000 FLOP | 2020-03-23 | Open weights (unrestricted) | Source ↗ |
| Incoder-6.7B | Facebook AI Research,University of Washington,University of California (UC) Berkeley,Carnegie Mellon University (CMU),Toyota Technological Institute at Chicago | 3,000,010,000,000,000,000,000 FLOP | 2023-04-09 | Open weights (non-commercial) | Source ↗ |
| AlphaFold 2 | DeepMind | 2,990,000,000,000,000,000,000 FLOP | 2020-11-30 | Open weights (unrestricted) | Source ↗ |
| CodeT5-large | Salesforce | 2,720,000,000,000,000,000,000 FLOP | 2022-07-05 | Open weights (unrestricted) | Source ↗ |
| StyleGAN3-R | Nvidia,Aalto University | 2,427,840,000,000,000,000,000 FLOP | 2021-06-21 | Open weights (non-commercial) | Source ↗ |
| InternImage | Shanghai AI Lab,Tsinghua University,Nanjing University,SenseTime,Chinese University of Hong Kong (CUHK) | 2,408,000,000,000,000,000,000 FLOP | 2022-11-10 | Open weights (unrestricted) | Source ↗ |
| ALBERT-xxlarge | Toyota Technological Institute at Chicago,Google | 2,390,000,000,000,000,000,000 FLOP | 2020-02-09 | Open weights (unrestricted) | Source ↗ |
| VILA1.5-13B | Nvidia,Massachusetts Institute of Technology (MIT) | 2,300,313,600,000,000,000,000 FLOP | 2024-05-03 | Open weights (non-commercial) | Source ↗ |
| VILA-13B | Nvidia,Massachusetts Institute of Technology (MIT) | 2,300,313,600,000,000,000,000 FLOP | 2023-12-12 | Open weights (non-commercial) | Source ↗ |
| NASv3 (CIFAR-10) | Google Brain | 2,200,000,000,000,000,000,000 FLOP | 2016-11-05 | — | Source ↗ |
| GPT-2 (1.5B) | OpenAI | 1,920,000,000,001,000,000,000 FLOP | 2019-02-14 | Open weights (unrestricted) | Source ↗ |
| EMDR | Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),McGill University,DeepMind | 1,910,000,000,000,000,000,000 FLOP | 2021-06-09 | Open weights (unrestricted) | Source ↗ |
| AlphaGo Lee | DeepMind | 1,900,000,000,000,000,000,000 FLOP | 2016-01-27 | Unreleased | Source ↗ |
| HyenaDNA | Stanford University,Harvard University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),University of Montreal / Université de Montréal | 1,811,000,000,000,000,000,000 FLOP | 2023-06-27 | Open weights (restricted use) | Source ↗ |
| StyleGAN3-T | Nvidia,Aalto University | 1,702,080,000,000,000,000,000 FLOP | 2021-06-21 | Open weights (non-commercial) | Source ↗ |
| KEPLER | Tsinghua University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),HEC,CIFAR AI Research,Princeton University,University of Montreal / Université de Montréal | 1,660,000,000,000,000,000,000 FLOP | 2020-11-23 | Unreleased | Source ↗ |
| CodeT5-base | Salesforce,Nanyang Technological University | 1,560,000,000,000,000,000,000 FLOP | 2021-11-01 | Open weights (unrestricted) | Source ↗ |
| German ELECTRA Large | deepset,Bayerische Staatsbibliothek Muenchen | 1,428,295,680,000,000,000,000 FLOP | 2020-10-21 | Open weights (unrestricted) | Source ↗ |
| GenSLM | University of Chicago,Nvidia,Harvard University,Cerebras Systems,Technical University of Munich,California Institute of Technology | 1,420,000,000,000,000,000,000 FLOP | 2022-10-11 | Open weights (unrestricted) | Source ↗ |
| BLIP-2 (Q-Former) | Salesforce Research | 1,200,000,000,010,000,000,000 FLOP | 2023-01-30 | Open weights (unrestricted) | Source ↗ |
| Megatron-LM (1.2B) | Nvidia | 1,130,400,000,000,000,000,000 FLOP | 2019-09-17 | Unreleased | Source ↗ |
| PolyCoder | Carnegie Mellon University (CMU) | 1,100,000,000,000,000,000,000 FLOP | 2022-02-26 | Open weights (unrestricted) | Source ↗ |
Data: Epoch AI · CC BY · View source · Automatically synced; not an exhaustive global sample