Model profile
Magistral Medium
Mistral AIdeveloper
2025-06-10release date
#154 / 267overall rank
6eval lineages
Evidence summary
Magistral Medium has an estimated overall rank of #154; its 90% source-sensitivity interval is #67–#245. Its behavior-only rank is #143; company governance moves the combined estimate to #154. Published evidence spans 6 evals and 6 of 7 behavior components. Its strongest relative result is FlagEval Safety and Values (a3_qualified_rate, #3 of 18); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #241 of 260).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗magistral-medium
- Official model documentation ↗Family-level model document · mistral · first party
- Release source ↗curated outward evidence
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #68 / 311 | 0.5947 | ↓ lower | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #132 / 260 | 15.76 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #200 / 260 | 83.17 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #241 / 260 | 33.33 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #202 / 258 | 91.82 | ↑ higher | Source ↗official | |
| FlagEval Safety and Valuesa1_qualified_rate | #12 / 18 | 79.01 | ↑ higher | Source ↗official | |
| FlagEval Safety and Valuesa2_qualified_rate | #12 / 18 | 78.73 | ↑ higher | Source ↗official | |
| FlagEval Safety and Valuesa3_qualified_rate | #3 / 18 | 89.28 | ↑ higher | Source ↗official | |
| FlagEval Safety and Valuesa4_qualified_rate | #7 / 18 | 89.15 | ↑ higher | Source ↗official | |
| FlagEval Safety and Valuesa5_qualified_rate | #3 / 18 | 74.27 | ↑ higher | Source ↗official | |
| HUMAINE Trust, Ethics and Safetytrust_ethics_safety_score | #36 / 54 | 26.12 | ↑ higher | Source ↗official | |
| MASKlying_probability_pct | #31 / 53 | 43.5 | ↓ lower | Source ↗official | |
| PHAREbias_resistance_diagnostic | #25 / 66 | 0.5075 | ↑ higher | Source ↗official | |
| PHAREhallucination_resistance_diagnostic | #49 / 70 | 0.6819 | ↑ higher | Source ↗official | |
| PHAREharm_resistance_diagnostic | #60 / 70 | 0.8452 | ↑ higher | Source ↗official | |
| PHAREjailbreak_resistance_diagnostic | #41 / 67 | 0.4561 | ↑ higher | Source ↗official |