Model profile
Phi 3.5 MoE Instruct
Evidence summary
Safety. Phi 3.5 MoE Instruct has an estimated Safety rank of #45; its 90% source-sensitivity interval is #8–#204. Its behavior-only rank is #35; company governance moves the combined estimate to #45. Published Safety evidence spans 5 eval lineages and 6 of 7 components. Its strongest relative result is PandaBench JBB direct-request panel (safety_rate, #1 of 46); its weakest is Microsoft Phi Safety Panels (harmful_summarization, #8 of 10).
Freedom. Phi 3.5 MoE Instruct has an estimated Freedom rank of #623; its 90% source-sensitivity interval is #473–#637. Published Freedom evidence spans 4 eval lineages and 1 of 1 components. Its strongest relative result is Microsoft Phi Safety Panels (harmful_summarization, #3 of 10); its weakest is Microsoft Phi Safety Panels (third_party_harm, #10 of 10).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗microsoft/Phi-3.5-MoE-instruct
- Model card ↗Exact model document · Microsoft · first party
- Release source ↗direct research preview identity date
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 |
|---|---|---|---|---|
| AILuminate General Purpose AI Chatchild_sexual_exploitation | #5 / 32 | ↓8.2 | Source ↗official | |
| AILuminate General Purpose AI Chatdefamation | #5 / 32 | ↓3.9 | Source ↗official | |
| AILuminate General Purpose AI Chathate | #5 / 32 | ↓2.5 | Source ↗official | |
| AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne | #4 / 32 | ↓6.7 | Source ↗official | |
| AILuminate General Purpose AI Chatintellectual_property | #6 / 32 | ↓4.5 | Source ↗official | |
| AILuminate General Purpose AI Chatnon_violent_crimes | #3 / 32 | ↓4.5 | Source ↗official | |
| AILuminate General Purpose AI Chatprivacy | #5 / 32 | ↓4.5 | Source ↗official | |
| AILuminate General Purpose AI Chatsex_related_crimes | #3 / 32 | ↓3.7 | Source ↗official | |
| AILuminate General Purpose AI Chatsexual_content | #4 / 32 | ↓4.7 | Source ↗official | |
| AILuminate General Purpose AI Chatspecialized_advice | #4 / 32 | ↓6.9 | Source ↗official | |
| AILuminate General Purpose AI Chatsuicide_self_harm | #4 / 32 | ↓3.7 | Source ↗official | |
| AILuminate General Purpose AI Chatviolent_crimes | #3 / 32 | ↓5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #152 / 270 | ↑14.99 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #134 / 270 | ↑87.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #124 / 270 | ↑76.11 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #59 / 268 | ↑98.32 | Source ↗official | |
| Large-scale Moral Machine experiment on LLMshuman_choice_distance | #16 / 39 | ↓0.9383 | Source ↗official | |
| Microsoft Phi Safety Panelsharmful_continuation | #2 / 10 | ↓0.004 | Source ↗official | |
| Microsoft Phi Safety Panelsharmful_summarization | #8 / 10 | ↓0.12 | Source ↗official | |
| Microsoft Phi Safety Panelsjailbreak | #2 / 10 | ↓0.11 | Source ↗official | |
| Microsoft Phi Safety Panelsthird_party_harm | #1 / 10 | ↓0.152 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #1 / 46 | ↑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 |
|---|---|---|---|---|
| AILuminate General Purpose AI Chatchild_sexual_exploitation | #28 / 32 | ↑8.2 | Source ↗official | |
| AILuminate General Purpose AI Chatdefamation | #28 / 32 | ↑3.9 | Source ↗official | |
| AILuminate General Purpose AI Chathate | #28 / 32 | ↑2.5 | Source ↗official | |
| AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne | #29 / 32 | ↑6.7 | Source ↗official | |
| AILuminate General Purpose AI Chatintellectual_property | #27 / 32 | ↑4.5 | Source ↗official | |
| AILuminate General Purpose AI Chatnon_violent_crimes | #30 / 32 | ↑4.5 | Source ↗official | |
| AILuminate General Purpose AI Chatprivacy | #28 / 32 | ↑4.5 | Source ↗official | |
| AILuminate General Purpose AI Chatsex_related_crimes | #30 / 32 | ↑3.7 | Source ↗official | |
| AILuminate General Purpose AI Chatsexual_content | #29 / 32 | ↑4.7 | Source ↗official | |
| AILuminate General Purpose AI Chatspecialized_advice | #29 / 32 | ↑6.9 | Source ↗official | |
| AILuminate General Purpose AI Chatsuicide_self_harm | #29 / 32 | ↑3.7 | Source ↗official | |
| AILuminate General Purpose AI Chatviolent_crimes | #30 / 32 | ↑5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #133 / 270 | ↓87.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #145 / 270 | ↓76.11 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #208 / 268 | ↓98.32 | Source ↗official | |
| Microsoft Phi Safety Panelsharmful_continuation | #9 / 10 | ↑0.004 | Source ↗official | |
| Microsoft Phi Safety Panelsharmful_summarization | #3 / 10 | ↑0.12 | Source ↗official | |
| Microsoft Phi Safety Panelsjailbreak | #9 / 10 | ↑0.11 | Source ↗official | |
| Microsoft Phi Safety Panelsthird_party_harm | #10 / 10 | ↑0.152 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #35 / 46 | ↓1 | Source ↗official |
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.