Model reference · open weights
Llama-Guard-3 is an open-weight language model from Meta.
Summary of the meta-llama/Llama-Guard-3-8B model card, 2026-10-01. The estimate below is for another build of the family.
What it is
| Released by | Meta |
|---|---|
| Released | 2024-07-22 |
| Parameters | 8.0B |
Builds
| Build | Params | Precision | Weights | Smallest setup |
|---|---|---|---|---|
| Llama-Guard-3-8B ↗ | 8.0B | BF16 | 16.1 GB | RTX 4090 24 GB |
| Llama-Guard-3-1B ↗ | 1.5B | BF16 | 3.0 GB | RTX 3060 12 GB |
| Llama-Guard-3-8B-INT8 ↗ | 8.0B | INT8 | 9.1 GB | RTX 3060 12 GB |
| Llama-Guard-3-1B-INT4 (above) ↗ | — | INT4 | — | — |
How it works