Model profile
Qwen2 72B Instruct
Evidence summary
Safety. Qwen2 72B Instruct has an estimated Safety rank of #102; its 90% source-sensitivity interval is #33–#189. Its behavior-only rank is #91; company governance moves the combined estimate to #102. Published Safety evidence spans 10 eval lineages and 5 of 7 components. Its strongest relative result is PandaBench JBB direct-request panel (safety_rate, #1 of 46); its weakest is AIRBench 2024 Safety Scenarios (safety_scenarios, #48 of 80).
Freedom. Qwen2 72B Instruct has an estimated Freedom rank of #411; its 90% source-sensitivity interval is #235–#630. Published Freedom evidence spans 8 eval lineages and 1 of 1 components. Its strongest relative result is HELM Safety (xstest, #27 of 80); its weakest is Qwen2 Safety Panel (fraud, #3 of 3).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗qwen2-72b-instruct
- Hugging Face ↗Qwen/Qwen2-72B-Instruct
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
Safety evals
Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AIRBench 2024 Safety Scenariossafety_scenarios | #48 / 80 | ↑0.621 | Source ↗official | |
| CASE-Benchagreement_accuracy | #2 / 7 | ↑84.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #141 / 270 | ↑15.76 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #77 / 270 | ↑90.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #84 / 270 | ↑87.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #28 / 268 | ↑99.36 | Source ↗official | |
| HELM Safetyanthropic_red_team | #32 / 80 | ↑0.991 | Source ↗official | |
| HELM Safetybbq | #26 / 80 | ↑0.951 | Source ↗official | |
| HELM Safetyharmbench | #38 / 80 | ↑0.768 | Source ↗official | |
| HELM Safetysimple_safety_tests | #43 / 80 | ↑0.985 | Source ↗official | |
| HELM Safetyxstest | #27 / 80 | ↑0.969 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #1 / 46 | ↑1 | Source ↗official | |
| Qwen2 Safety Panelfraud | #1 / 3 | ↓2.41 | Source ↗official | |
| Qwen2 Safety Panelillegal | #1 / 3 | ↓0 | Source ↗official | |
| Qwen2 Safety Panelpornography | #1 / 3 | ↓22.91 | Source ↗official | |
| Qwen2 Safety Panelprivacy | #1 / 3 | ↓2.47 | Source ↗official |
Freedom evals
Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AIRBench 2024 Safety Scenariossafety_scenarios | #33 / 80 | ↓0.621 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #189 / 270 | ↓90.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #186 / 270 | ↓87.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #239 / 268 | ↓99.36 | Source ↗official | |
| HELM Safetyanthropic_red_team | #45 / 80 | ↓0.991 | Source ↗official | |
| HELM Safetyharmbench | #43 / 80 | ↓0.768 | Source ↗official | |
| HELM Safetysimple_safety_tests | #34 / 80 | ↓0.985 | Source ↗official | |
| HELM Safetyxstest | #27 / 80 | ↑0.969 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #35 / 46 | ↓1 | Source ↗official | |
| Qwen2 Safety Panelfraud | #3 / 3 | ↑2.41 | Source ↗official | |
| Qwen2 Safety Panelillegal | #2 / 3 | ↑0 | Source ↗official | |
| Qwen2 Safety Panelpornography | #3 / 3 | ↑22.91 | Source ↗official | |
| Qwen2 Safety Panelprivacy | #3 / 3 | ↑2.47 | Source ↗official |