Llama-Primus-Merged: benchmark results
Provider: Meta.
Aggregate Rating 1490 ± 49, rank #2461 of 10556 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AthenaBench - ATE | 33.8 | Accuracy (%) extracting ATT&CK techniques from threat report | 75 |
| AthenaBench | 43.7 | Combined score (%): mean of the six task scores, full benchm | 57.1 |
| AthenaBench - RCM | 56.6 | Accuracy (%) mapping vulnerability descriptions to their CWE | 44.4 |
| AthenaBench - CKT | 76.3 | Accuracy (%) on the CTI knowledge test (3,000 questions) | 42.9 |
| AthenaBench - RMS | 6.6 | F1 (%) of proposed ATT&CK mitigations against the reference | 37.5 |
| AthenaBench - TAA | 17 | Accuracy (%) attributing anonymised incident reports to a th | 31.2 |
| AthenaBench - VSP | 71.9 | Accuracy (%) = 1 - MAD/7.7 of predicted CVSS base scores ove | 22.2 |
| KaliBench | 16.3 | Exact Correct (%), Unrestricted mode: share of the 5,000 ver | 17.4 |
| RedSage-Bench | 74.81 | Macro Accuracy (imported) | 6.2 |
Interactive version: theaggregate.ai/model?slug=llama-primus-merged · How It Works · Data refreshed daily, snapshot 2026-11-17.