Model profile
Vicuna 13B
LMSYSdeveloper
2023-03-30release date
#189 / 267overall rank
3eval lineages
Evidence summary
Vicuna 13B has an estimated overall rank of #189; its 90% source-sensitivity interval is #120–#217. Its behavior-only rank is #192; company governance moves the combined estimate to #189. Published evidence spans 3 evals and 3 of 7 behavior components. Its strongest relative result is Fake Alignment (FINE) (multiple_choice_safe_decision_rate, #5 of 14); its weakest is SafetyBench (UB, #18 of 21).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗lmsys/vicuna-13b-v1.3
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
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 | Better | Source |
|---|---|---|---|---|---|
| Fake Alignment (FINE)multiple_choice_safe_decision_rate | #5 / 14 | 58.67 | ↑ higher | Source ↗official | |
| Fake Alignment (FINE)open_ended_safe_response_rate | #7 / 14 | 96 | ↑ higher | Source ↗official | |
| S-Evalbase_en_overall | #13 / 22 | 57.1 | ↑ higher | Source ↗official | |
| SafetyBenchEM | #14 / 21 | 65.9 | ↑ higher | Source ↗official | |
| SafetyBenchIA | #13 / 21 | 71.4 | ↑ higher | Source ↗official | |
| SafetyBenchMH | #14 / 21 | 77.5 | ↑ higher | Source ↗official | |
| SafetyBenchOFF | #8 / 21 | 68.4 | ↑ higher | Source ↗official | |
| SafetyBenchPH | #14 / 21 | 65.3 | ↑ higher | Source ↗official | |
| SafetyBenchPP | #12 / 21 | 75.4 | ↑ higher | Source ↗official | |
| SafetyBenchUB | #18 / 21 | 53 | ↑ higher | Source ↗official |