Closed-weight language models from Google DeepMind
12
Models
1 organisations
0%
Open weights
0 open, 12 closed
1
Frontier
By Epoch's flag
—
Largest traceable
No traceable figure
2024–2026
Span
0 with a traceable size
Organisation
Country
Records
12| Model | Organisation | Released | Parameters | Weights |
|---|---|---|---|---|
| Gemini 4 Carbon | Google DeepMind | 2026-10 | — | Closed |
| Gemini 3.8 Flash | Google DeepMind | 2026-09 | — | Closed |
| Gemini 3.5 Flash | Google DeepMind | 2026-05 | — | Closed |
| Gemini 3.0 Flash-lite | Google DeepMind | 2026-03 | — | Closed |
| Gemini 3 Deep Think | Google DeepMind | 2025-11 | — | Closed |
| Gemini Embedding | Google DeepMind | 2025-07 | — | Closed |
| Gemini 2.5 Pro (Jun 2025) | Google DeepMind | 2025-06 | — | Closed |
| Gemini 2.5 Pro (May 2025) | Google DeepMind | 2025-05 | — | Closed |
| Gemini 2.5 Pro (Mar 2025) | Google DeepMind | 2025-03 | — | Closed |
| Gemini 2.0 Pro | Google DeepMind | 2024-12 | — | Closed |
| Gemini-Exp-1114 | Google DeepMind | 2024-11 | — | Closed |
| Gemini 1.5 Pro | Google DeepMind | 2024-02 | — | Closed |
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.