Model profile
Qwen3.5 4B
Evidence summary
Safety. Qwen3.5 4B has an estimated Safety rank of #241; its 90% source-sensitivity interval is #40–#303. Its behavior-only rank is #236; company governance moves the combined estimate to #241. Published Safety evidence spans 3 eval lineages and 5 of 7 components. Its strongest relative result is VETO Misfired Alignment (misfired_alignment_rate_pct, #10 of 25); its weakest is AA-Omniscience (hallucination_rate, #232 of 345).
Freedom. Qwen3.5 4B has an estimated Freedom rank of #415; its 90% source-sensitivity interval is #164–#636. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is VETO Misfired Alignment (misfired_alignment_rate_pct, #10 of 25); its weakest is UGI Leaderboard — base-model willingness (willingness_adherence_score, #137 of 156).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗qwen3-5-4b
- Hugging Face ↗Qwen/Qwen3.5-4B
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗repository created at proxy
Finetunes
Direct finetunes linked to this canonical base model. Quantized and repackaged derivatives are excluded.
darkc0de/XORTRON.CriminalComputing.2026.4B.Instruct.NEXT ↗Direct non-quantized finetune documented by Hugging Face metadata
dealignai/Qwen3.5-VL-4B-JANG_4S-CRACK ↗Direct non-quantized finetune documented by Hugging Face metadata- MuXodious/Qwen3.5-4B-SOMPOA-heresy-v2 ↗Direct non-quantized finetune documented by Hugging Face metadata
- prithivMLmods/Qwen3.5-4B-Unredacted-MAX ↗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 |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #232 / 345 | ↓0.8655 | Source ↗official | |
| KIDBench Implicit Child Cueimplicit_child_cue_total_mean | #8 / 13 | ↑3.9 | Source ↗official | |
| VETO Misfired Alignmentmisfired_alignment_rate_pct | #10 / 25 | ↓7.9 | 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 |
|---|---|---|---|---|
| UGI Leaderboard — base-model willingnesswillingness_adherence_score | #137 / 156 | ↑0.75 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_direct_score | #99 / 156 | ↑2.5 | Source ↗official | |
| VETO Misfired Alignmentmisfired_alignment_rate_pct | #10 / 25 | ↓7.9 | 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.