# Winnow-12B Q8 benchmark scores

Other · model · https://benchgap.net/model/winnow-12b · data retrieved 2026-11-22

As of 2026-11-22, Winnow-12B Q8 (Other) has measured scores on 7 benchmarks and estimated scores on 5 more.

| Benchmark | Score | Source |
|---|---|---|
| [JevBench 1.4](https://benchgap.net/b/jev-bench14/current.md) | 55.6% | measured |
| [Decision Index Full Score](https://benchgap.net/b/decisionindex/current.md) | 46.6% | measured |
| [Decision Index Arts](https://benchgap.net/b/jevdecisionindex03arts/current.md) | 35.1% | measured |
| [Decision Index Knowledge](https://benchgap.net/b/jevdecisionindex03knowledge/current.md) | 31.2% | measured |
| [Decision Index Language](https://benchgap.net/b/jevdecisionindex03language/current.md) | 50.6% | measured |
| [Decision Index Retrieval](https://benchgap.net/b/jevdecisionindex03retrieval/current.md) | 54.6% | measured |
| [Decision Index Tools](https://benchgap.net/b/jevdecisionindex03tools/current.md) | 62.9% | measured |
| [HLE-MultiChoice (Decision Index)](https://benchgap.net/b/aa-decision-hle-multi-choice/current.md) | 11.0% | estimated ± 2.1 pp, high confidence |
| [Artificial Analysis Decision Index v1.0](https://benchgap.net/b/aa-decision-index/current.md) | 42.2% | estimated ± 3.2 pp, medium confidence |
| [Artificial Analysis Decision Index](https://benchgap.net/b/aa-decision-index-v2/current.md) | 57.1% | estimated ± 1.3 pp, medium confidence |
| [When2Call (Decision Index)](https://benchgap.net/b/aa-decision-when2-call/current.md) | 57.6% | estimated ± 3.0 pp, medium confidence |
| [WildGuardTest (Decision Index)](https://benchgap.net/b/aa-decision-wild-guard-test/current.md) | 90.7% | estimated ± 1.9 pp, high confidence |
