Model profile
Mistral Medium 3.1
Evidence summary
Mistral Medium 3.1 has an estimated overall rank of #209; its 90% source-sensitivity interval is #116–#246. Its behavior-only rank is #201; company governance moves the combined estimate to #209. Published evidence spans 5 evals and 6 of 7 behavior components. Its strongest relative result is UAVBench safety-critical decision recognition (ethical_safety_critical_accuracy, #6 of 27); its weakest is SpeciEval (sea_animal_4ns, #100 of 102).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗mistral-medium-3-1
- OpenRouter ↗mistralai/mistral-medium-3.1
- 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 | #163 / 311 | 0.813 | ↓ lower | Source ↗official | |
| Alignment Leaderboardcorrigibility | #17 / 24 | 4.117 | ↑ higher | Source ↗official | |
| Alignment Leaderboardhonesty | #20 / 24 | 3.212 | ↑ higher | Source ↗official | |
| Alignment Leaderboardnon_manipulation | #22 / 24 | 2.665 | ↑ higher | Source ↗official | |
| Alignment Leaderboardrobustness | #19 / 24 | 3.147 | ↑ higher | Source ↗official | |
| Alignment Leaderboardsafety | #23 / 24 | 2.962 | ↑ higher | Source ↗official | |
| Alignment Leaderboardscheming | #21 / 24 | 3.319 | ↑ higher | Source ↗official | |
| PHAREbias_resistance_diagnostic | #45 / 66 | 0.3966 | ↑ higher | Source ↗official | |
| PHAREhallucination_resistance_diagnostic | #46 / 70 | 0.6963 | ↑ higher | Source ↗official | |
| PHAREharm_resistance_diagnostic | #40 / 70 | 0.9232 | ↑ higher | Source ↗official | |
| PHAREjailbreak_resistance_diagnostic | #63 / 67 | 0.3083 | ↑ higher | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #33 / 102 | 6.88 | ↑ higher | Source ↗official | |
| SpeciEvalland_animal_4ns | #89 / 102 | 4.97 | ↓ lower | Source ↗official | |
| SpeciEvalsea_animal_4ns | #100 / 102 | 5.58 | ↓ lower | Source ↗official | |
| SpeciEvalspeciesism | #56 / 102 | 2.09 | ↓ lower | Source ↗official | |
| UAVBench safety-critical decision recognitionethical_safety_critical_accuracy | #6 / 27 | 0.75 | ↑ higher | 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.