Open-weight language models in 2021
16
Models
12 organisations
100%
Open weights
16 open, 0 closed
1
Frontier
By Epoch's flag
1.6T
Largest traceable
Switch
2021
Span
12 with a traceable size
Releases
by monthOrganisation
Country
Records
16| Model | Organisation | Released | Parameters | Weights |
|---|---|---|---|---|
| Contriever | Meta AI | 2021-12 | 110M * | Open |
| LongT5 | Google Research | 2021-12 | 3B | Open |
| XGLM-7.5B | Meta AI | 2021-12 | 8B | Open |
| CodeT5-base | Salesforce | 2021-11 | 220M * | Open |
| base LM+GNN+kNN | Shannon.AI | 2021-10 | 274M | Open |
| S4 | Stanford University | 2021-10 | 249M * | Open |
| PLATO-XL | Baidu | 2021-09 | 11B | Open |
| XLMR-XXL | Facebook AI Research | 2021-08 | 11B | Open |
| ERNIE 3.0 | Baidu | 2021-07 | 10B | Open |
| DeBERTa | Microsoft | 2021-06 | 2B | Open |
| EMDR | Mila - Quebec AI (originally Montreal Institute for Learning Algorithms) | 2021-06 | 440M | Open |
| ByT5-XXL | 2021-05 | 13B * | Open | |
| Generative BST | Facebook AI Research | 2021-03 | 9B | Open |
| Unicorn | Allen Institute for AI | 2021-03 | 11B | Open |
| SRU++ Large | ASAPP | 2021-02 | 234M | Open |
| Switch | 2021-01 | 1.6T | Open |
An asterisk marks a parameter figure the source does not consider traceable to the people who built the model. That judgement is the source’s and it has false negatives: Kimi K3 is marked speculative at 2.8T while Moonshot’s own model card states the figure. Why that distinction carries the whole exhibit.
Counts are of models Epoch AI records as notable, not of every model released. The most recent month is always incomplete, because a model enters the dataset when it is reviewed rather than when it ships.
Epoch AI, 'Data on AI Models'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/data/ai-models-documentation' Used under CC BY 4.0.