Model profile
Qwen2.5 3B Instruct
Alibabadeveloper
2024-09-17release date
#226 / 267overall rank
3eval lineages
Evidence summary
Qwen2.5 3B Instruct has an estimated overall rank of #226; its 90% source-sensitivity interval is #54–#265. Its behavior-only rank is #222; company governance moves the combined estimate to #226. Published evidence spans 3 evals and 3 of 7 behavior components. Its strongest relative result is Open LLM Safety Index (jailbreakbench_safety_rate, #4 of 21); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #219 of 260).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗Qwen/Qwen2.5-3B-Instruct
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗repository created at proxy
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 |
|---|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #219 / 260 | 10.59 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #155 / 260 | 87 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #218 / 260 | 41.67 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #165 / 258 | 95.09 | ↑ higher | Source ↗official | |
| Open LLM Safety Indexjailbreakbench_safety_rate | #4 / 21 | 0.6 | ↑ higher | Source ↗official | |
| Open LLM Safety Indexstrongreject_safety_rate | #8 / 21 | 0.3333 | ↑ higher | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #14 / 46 | 0.99 | ↑ higher | Source ↗official |