Model profile
DeepSeek V4 Flash 0731
DeepSeekdeveloper
2026-07-31release date
#191 / 317overall rank
4eval lineages
1discovery sources
Evidence summary
DeepSeek V4 Flash 0731 has an estimated overall rank of #191; its 90% source-sensitivity interval is #75–#288. Its behavior-only rank is #181; company governance moves the combined estimate to #191. Published evidence spans 4 evals and 2 of 7 behavior components. Its strongest relative result is SpeciEval (belief_animal_sentience, #1 of 113); its weakest is AA-Omniscience (hallucination_rate, #278 of 331).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗fe4c0848-e284-4e52-a79d-cdc28392f1a9
- Hugging Face ↗deepseek-ai/DeepSeek-V4-Flash-0731
- OpenRouter ↗deepseek/deepseek-v4-flash-0731
- Official model documentation ↗Family-level model document · deepseek · first party
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 | Source |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #278 / 331 | ↓0.917 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #53 / 106 | ↑0.29 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #1 / 113 | ↑7 | Source ↗official | |
| SpeciEvalland_animal_4ns | #93 / 113 | ↓4.85 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #87 / 113 | ↓4.95 | Source ↗official | |
| SpeciEvalspeciesism | #7 / 113 | ↓1.23 | Source ↗official | |
| TACbase_welfare_rate | #61 / 79 | ↑20.51 | Source ↗self run |