Model profile
DeepSeek V4 Flash Vision
Evidence summary
Safety. DeepSeek V4 Flash Vision does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 1 of 7 components. Its strongest relative result is AA-Omniscience (hallucination_rate, #290 of 345); its weakest is AA-Omniscience (hallucination_rate, #290 of 345).
Freedom. DeepSeek V4 Flash Vision has an estimated Freedom rank of #265; its 90% source-sensitivity interval is #110–#481. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SpeechMap model completion (complete_pct, #51 of 181); its weakest is SpeechMap model completion (complete_pct, #51 of 181).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗deepseek-v4-flash-vision
- OpenRouter ↗deepseek/deepseek-v4-flash-vision-exp
- Official model documentation ↗Family-level model document · deepseek · first party
- Release source ↗curated outward evidence
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 | #290 / 345 | ↓0.9148 | 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 |
|---|---|---|---|---|
| SpeechMap model completioncomplete_pct | #51 / 181 | ↑71.7 | Source ↗official |