architectures · google
Decoder-only transformer with interleaved local (sliding window) and global attention. Multimodal via SigLIP vision encoder. 128K context.
| Weight file SHA-256 | cf5ce22f4d925aa6befbae81e02fdf3ab3bbfb42e794ad2e176364bfb53d75ab |
|---|
| Publisher | |
|---|---|
| Origin | United States |
| Licence | Gemma-TOS |
| Modality | Text/Image |
| Pulls recorded here | 44,001 |
Registry record as of 2026-06-06. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.