Model profile
Qwen2.5 0.5B Instruct
Evidence summary
Safety. Qwen2.5 0.5B Instruct has an estimated Safety rank of #330; its 90% source-sensitivity interval is #304–#333. Published Safety evidence spans 4 eval lineages and 3 of 7 components. Its strongest relative result is Open LLM Safety Index (jailbreakbench_safety_rate, #12 of 21); its weakest is M3-SafetyBench (overall_score, #19 of 19).
Freedom. Qwen2.5 0.5B Instruct has an estimated Freedom rank of #16; its 90% source-sensitivity interval is #17–#178. Published Freedom evidence spans 3 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #12 of 248); its weakest is Open LLM Safety Index (jailbreakbench_safety_rate, #9 of 21).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗Qwen/Qwen2.5-0.5B-Instruct
- Official model page ↗Family-level model document · alibaba · first party
- Release source ↗direct research preview identity date
Finetunes
Direct finetunes linked to this canonical base model. Quantized and repackaged derivatives are excluded.
huihui-ai/Qwen2.5-0.5B-Instruct-abliterated-v3 ↗Direct non-quantized finetune documented by Hugging Face metadata- huihui-ai/Qwen2.5-0.5B-Instruct-CensorTune ↗Direct non-quantized finetune documented by Hugging Face metadata
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 |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #242 / 248 | ↑3.88 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #191 / 248 | ↑81.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #236 / 248 | ↑31.11 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #181 / 246 | ↑93.36 | Source ↗official | |
| M3-SafetyBenchoverall_score | #19 / 19 | ↑69.63 | Source ↗official | |
| Open LLM Safety Indexjailbreakbench_safety_rate | #12 / 21 | ↑0.3333 | Source ↗official | |
| Open LLM Safety Indexstrongreject_safety_rate | #16 / 21 | ↑0.0667 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #40 / 46 | ↑0.94 | 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 |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #57 / 248 | ↓81.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #12 / 248 | ↓31.11 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #66 / 246 | ↓93.36 | Source ↗official | |
| Open LLM Safety Indexjailbreakbench_safety_rate | #9 / 21 | ↓0.3333 | Source ↗official | |
| Open LLM Safety Indexstrongreject_safety_rate | #4 / 21 | ↓0.0667 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #6 / 46 | ↓0.94 | Source ↗official |