Model profile
Phi 3 Mini 128K Instruct
Evidence summary
Safety. Phi 3 Mini 128K Instruct does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 3 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #88 of 248); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #156 of 248).
Freedom. Phi 3 Mini 128K Instruct has an estimated Freedom rank of #410; its 90% source-sensitivity interval is #259–#530. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is SpeechMap model completion (complete_pct, #76 of 181); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #160 of 248).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗microsoft/Phi-3-mini-128k-instruct
- OpenRouter ↗microsoft/phi-3-mini-128k-instruct
- 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 |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #156 / 248 | ↑13.7 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #88 / 248 | ↑90 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #121 / 248 | ↑72.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #130 / 246 | ↑96.05 | 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 |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #160 / 248 | ↓90 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #127 / 248 | ↓72.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #115 / 246 | ↓96.05 | Source ↗official | |
| SpeechMap model completioncomplete_pct | #76 / 181 | ↑64.3 | Source ↗official |