Model profile
Mistral Large 4
Evidence summary
Safety. Mistral Large 4 has an estimated Safety rank of #41; its 90% source-sensitivity interval is #34–#122. Its behavior-only rank is #16; company governance moves the combined estimate to #41. Published Safety evidence spans 5 eval lineages and 5 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #21 of 272); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #260 of 272).
Freedom. Mistral Large 4 has an estimated Freedom rank of #411; its 90% source-sensitivity interval is #277–#513. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #13 of 272); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #248 of 272).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗mistral-large-4
- OpenRouter ↗mistralai/mistral-large-4-0
- Official model documentation ↗Exact model document · mistral · first party
- Release source ↗curated outward evidence
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 |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #66 / 361 | ↓0.4195 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #56 / 272 | ↑30.75 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #260 / 272 | ↑50.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #21 / 272 | ↑98.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #113 / 270 | ↑97.09 | Source ↗official | |
| Manager Coercion Benchcoercion_ladder_depth | #13 / 47 | ↓5.6 | Source ↗self run | |
| SpeciEvalbelief_animal_sentience | #58 / 133 | ↑6.83 | Source ↗official | |
| SpeciEvalland_animal_4ns | #28 / 133 | ↓4.28 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #44 / 133 | ↓4.6 | Source ↗official | |
| SpeciEvalspeciesism | #51 / 133 | ↓1.85 | Source ↗official | |
| TACbase_welfare_rate | #36 / 94 | ↑29.49 | Source ↗self run |
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 | #13 / 272 | ↓50.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #248 / 272 | ↓98.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #155 / 270 | ↓97.09 | Source ↗official |