Model profile
DeepSeek V4 Pro 0813
DeepSeekdeveloper
2026-08-13release date
Not rankedSafety rank
Not rankedFreedom rank
Evidence summary
Safety. DeepSeek V4 Pro 0813 does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 1 of 7 components. Its strongest relative result is BullshitBench v2 (clear_pushback_rate, #60 of 122); its weakest is AA-Omniscience (hallucination_rate, #312 of 358).
Freedom. DeepSeek V4 Pro 0813 does not meet the evidence gate for a Freedom rank.
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗d4f5b250-be37-4801-85e6-32fda4179337
- Hugging Face ↗deepseek-ai/DeepSeek-V4-Pro-0813
- Model card ↗Exact model document · DeepSeek · official repository
- Release source ↗repository created at proxy
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 | #312 / 358 | ↓0.9194 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #60 / 122 | ↑0.335 | Source ↗official |
Freedom evals
No published sub-eval result contributes to this model’s Freedom profile.