Model profile
Qwen 2.5 1.5B Instruct
Evidence summary
Qwen 2.5 1.5B Instruct has an estimated overall rank of #224; its 90% source-sensitivity interval is #95–#260. Its behavior-only rank is #221; company governance moves the combined estimate to #224. Published evidence spans 5 evals and 3 of 7 behavior components. Its strongest relative result is PandaBench JBB direct-request panel (safety_rate, #1 of 46); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #241 of 260).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗Qwen/Qwen2.5-1.5B-Instruct
- 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 |
|---|---|---|---|---|---|
| DSPSafeBenchscore | #9 / 12 | 71.26 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #241 / 260 | 8.01 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #180 / 260 | 85.17 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #181 / 260 | 54.44 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #128 / 258 | 96.36 | ↑ higher | Source ↗official | |
| M3-SafetyBenchoverall_score | #14 / 19 | 85.87 | ↑ 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 | #1 / 46 | 1 | ↑ higher | Source ↗official |
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.