Model profile
Qwen3 Next 80B A3B Instruct
Evidence summary
Safety. Qwen3 Next 80B A3B Instruct has an estimated Safety rank of #295; its 90% source-sensitivity interval is #113–#329. Its behavior-only rank is #292; company governance moves the combined estimate to #295. Published Safety evidence spans 4 eval lineages and 2 of 7 components. Its strongest relative result is AgentDrive Safety Compliance (scr, #6 of 48); its weakest is Arena Factuality — Text Arena (factuality-only weighting) (factuality_bt_rating, #104 of 111).
Freedom. Qwen3 Next 80B A3B Instruct has an estimated Freedom rank of #336; its 90% source-sensitivity interval is #126–#473. Published Freedom evidence spans 3 eval lineages and 1 of 1 components. Its strongest relative result is Human Pathogen Capabilities Test (HPCT) — overall refusal (refusal_rate_pct, #1 of 69); its weakest is SpeechMap model completion (complete_pct, #112 of 181).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗qwen3-next-80b-a3b-instruct
- Hugging Face ↗Qwen/Qwen3-Next-80B-A3B-Instruct
- NVIDIA ↗qc69jvmznzxy/qwen3-next-80b-a3b-instruct
- OpenRouter ↗qwen/qwen3-next-80b-a3b-instruct
- 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.
- blascotobasco/Qwen3-Next-352E-Abliterated-Instruct ↗Direct non-quantized finetune documented by Hugging Face metadata
- temaq-org/Tema_Q-X3 ↗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 | #304 / 345 | ↓0.9275 | Source ↗official | |
| AgentDrive Safety Compliancescr | #6 / 48 | ↑95 | Source ↗official | |
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #104 / 111 | ↑1392.0 | Source ↗official | |
| Human Pathogen Capabilities Test (HPCT) — overall refusalrefusal_rate_pct | #27 / 69 | ↑0 | 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 |
|---|---|---|---|---|
| Human Pathogen Capabilities Test (HPCT) — overall refusalrefusal_rate_pct | #1 / 69 | ↓0 | Source ↗official | |
| SpeechMap model completioncomplete_pct | #112 / 181 | ↑49.1 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_adherence_score | #79 / 156 | ↑1.5 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_direct_score | #46 / 156 | ↑4 | 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.