Model profile
Phi 4 Reasoning Plus
Evidence summary
Safety. Phi 4 Reasoning Plus has an estimated Safety rank of #137; its 90% source-sensitivity interval is #28–#248. Its behavior-only rank is #140; company governance moves the combined estimate to #137. Published Safety evidence spans 3 eval lineages and 5 of 7 components. Its strongest relative result is AgentDrive Safety Compliance (scr, #6 of 48); its weakest is FlagEval Safety and Values (a2_qualified_rate, #13 of 18).
Freedom. Phi 4 Reasoning Plus has an estimated Freedom rank of #401; its 90% source-sensitivity interval is #235–#560. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SpeechMap model completion (complete_pct, #67 of 181); its weakest is SpeechMap model completion (complete_pct, #67 of 181).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗microsoft/Phi-4-reasoning-plus
- OpenRouter ↗microsoft/phi-4-reasoning-plus
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · 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 |
|---|---|---|---|---|
| AgentDrive Safety Compliancescr | #6 / 48 | ↑95 | Source ↗official | |
| FlagEval Safety and Valuesa1_qualified_rate | #7 / 18 | ↑81.55 | Source ↗official | |
| FlagEval Safety and Valuesa2_qualified_rate | #13 / 18 | ↑78.31 | Source ↗official | |
| FlagEval Safety and Valuesa3_qualified_rate | #6 / 18 | ↑87.76 | Source ↗official | |
| FlagEval Safety and Valuesa4_qualified_rate | #11 / 18 | ↑87.95 | Source ↗official | |
| FlagEval Safety and Valuesa5_qualified_rate | #13 / 18 | ↑69.65 | Source ↗official | |
| UAVBench safety-critical decision recognitionethical_safety_critical_accuracy | #7 / 27 | ↑0.735 | 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 | #67 / 181 | ↑67.3 | Source ↗official |