models · meta-llama
Meta Llama 3.2 largest multimodal variant. 90B vision-language instruct model for complex image understanding and document analysis.
| Weight file SHA-256 | 3f908d85029f1ea999d701f63b857ed87c399753d6ed49a448e43df89fa180d6 |
|---|---|
| Method | hf-published-sha256 |
| Publisher | meta-llama |
|---|---|
| Origin | United States |
| Licence | LLaMA-3.2 |
| Publisher’s own licence statement | llama3.2 |
| Parameters | 88,593,355,323 (88.6B) |
| Context window | 128K |
| Native precision | BF16 |
| Pulls recorded here | 360,056 |
| Downloads reported upstream | 150,548 |
| MMLU-Pro | 67.1 |
|---|---|
| GPQA Diamond | 43.2 |
| MATH-500 | 62.9 |
| LiveCodeBench | 21.4 |
| Humanity's Last Exam | 4.5 |
Reported by a third party and recorded here, not re-run by us. Source: artificialanalysis.ai/api/v2.
Registry record as of 2026-10-14. 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.