Model profile
Qwen3 4B
2025-04-27release date
2eval lineages
1discovery sources
Evidence summary
Published evidence spans 2 evals and 3 of 7 behavior components. Its strongest relative result is VETO Misfired Alignment (misfired_alignment_rate_pct, #10 of 25); its weakest is PandaBench JBB direct-request panel (safety_rate, #28 of 46).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗Qwen/Qwen3-4B
- OpenRouter ↗qwen/qwen3-4b
- 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 |
|---|---|---|---|---|---|
| PandaBench JBB direct-request panelsafety_rate | #28 / 46 | 0.97 | ↑ higher | Source ↗official | |
| VETO Misfired Alignmentmisfired_alignment_rate_pct | #10 / 25 | 7.9 | ↓ lower | 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.