Model profile
DeepSeek R1 Zero
Evidence summary
Safety. DeepSeek R1 Zero 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 Reward Hacking Benchmark (integrity_score, #13 of 13); its weakest is Reward Hacking Benchmark (integrity_score, #13 of 13).
Freedom. DeepSeek R1 Zero has an estimated Freedom rank of #43; its 90% source-sensitivity interval is #8–#410. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SpeechMap model completion (complete_pct, #9 of 181); its weakest is SpeechMap model completion (complete_pct, #9 of 181).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗deepseek-ai/DeepSeek-R1-Zero
- OpenRouter ↗deepseek/deepseek-r1-zero
- Official model documentation ↗Family-level model document · deepseek · first party
- 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 |
|---|---|---|---|---|
| Reward Hacking Benchmarkintegrity_score | #13 / 13 | ↑86.1 | 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 | #9 / 181 | ↑91.6 | Source ↗official |