Model profile
Deepseek V3.2 Speciale
Evidence summary
Deepseek V3.2 Speciale has an estimated overall rank of #249; its 90% source-sensitivity interval is #131–#265. Its behavior-only rank is #246; company governance moves the combined estimate to #249. Published evidence spans 3 evals and 3 of 7 behavior components. Its strongest relative result is Cisco AI Defense Rolling Single-Turn Leaderboard (single_turn_attack_success_rate, #71 of 105); its weakest is SABER (overall_safety_rate, #12 of 13).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗deepseek-v3-2-speciale
- Hugging Face ↗deepseek-ai/DeepSeek-V3.2-Speciale
- OpenRouter ↗deepseek/deepseek-v3.2-speciale
- 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 |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #242 / 311 | 0.8894 | ↓ lower | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #71 / 105 | 54.99 | ↓ lower | Source ↗official | |
| SABERoverall_safety_rate | #12 / 13 | 20.42 | ↑ higher | Source ↗official | |
| SABERscenario_a_safety_rate | #9 / 13 | 26.75 | ↑ higher | Source ↗official | |
| SABERscenario_b_safety_rate | #11 / 13 | 25.16 | ↑ higher | Source ↗official | |
| SABERscenario_c_safety_rate | #12 / 13 | 9.77 | ↑ higher | 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.