Model profile
Llama Guard 4 12B
1eval lineages
Evidence summary
Published evidence spans 1 evals and 1 of 7 behavior components. Its strongest relative result is AIMS Safety-Classifier Competence (average_harmful_f1, #11 of 11); its weakest is AIMS Safety-Classifier Competence (average_harmful_f1, #11 of 11).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗meta-llama/Llama-Guard-4-12B
- NVIDIA ↗qc69jvmznzxy/llama-guard-4-12b
- OpenRouter ↗meta-llama/llama-guard-4-12b
- Official model page ↗Family-level model document · meta · first party
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AIMS Safety-Classifier Competenceaverage_harmful_f1 | #11 / 11 | ↑0.691 | Source ↗official |