Model profile
Qwen3.8 27B
Evidence summary
Safety. Qwen3.8 27B has an estimated Safety rank of #155; its 90% source-sensitivity interval is #13–#290. Its behavior-only rank is #149; company governance moves the combined estimate to #155. Published Safety evidence spans 4 eval lineages and 3 of 7 components. Its strongest relative result is AA-Omniscience (hallucination_rate, #7 of 345); its weakest is TAC (base_welfare_rate, #79 of 87).
Freedom. Qwen3.8 27B has an estimated Freedom rank of #611; its 90% source-sensitivity interval is #397–#636. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SpeechMap model completion (complete_pct, #147 of 181); its weakest is SpeechMap model completion (complete_pct, #147 of 181).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗qwen3-8-27b
- Hugging Face ↗Qwen/Qwen3.8-27B
- OpenRouter ↗qwen/qwen3.8-27b
- Official model page ↗Family-level model document · alibaba · first party
Finetunes
Direct finetunes linked to this canonical base model. Quantized and repackaged derivatives are excluded.
Blackfrost-AI/Qwen3.8-27B-ABLITERATED-BF16 ↗Direct non-quantized finetune documented by Hugging Face metadata
heretic-org/Qwen3.8-27B-heretic-ara ↗Direct non-quantized finetune documented by Hugging Face metadata
huihui-ai/Huihui-Qwen3.8-27B-abliterated ↗Direct non-quantized finetune documented by Hugging Face metadata
orcarouter/Qwen3.8-27B-Uncensored ↗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 | #7 / 345 | ↓0.1809 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #28 / 117 | ↑0.54 | Source ↗official | |
| Manager Coercion Benchcoercion_ladder_depth | #8 / 37 | ↓5.567 | Source ↗self run | |
| TACbase_welfare_rate | #79 / 87 | ↑16.67 | Source ↗self run |
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 |
|---|---|---|---|---|
| SpeechMap model completioncomplete_pct | #147 / 181 | ↑38.5 | Source ↗official |