Model profile
Minimax M2
Evidence summary
Minimax M2 has an estimated overall rank of #174; its 90% source-sensitivity interval is #55–#275. Its behavior-only rank is #175; company governance moves the combined estimate to #174. Published evidence spans 6 evals and 6 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #14 of 241); its weakest is SpeciEval (sea_animal_4ns, #105 of 113).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗minimax-m2
- Hugging Face ↗MiniMaxAI/MiniMax-M2
- OpenRouter ↗minimax/minimax-m2
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
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 | Source |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #264 / 328 | ↓0.909 | Source ↗official | |
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #96 / 112 | ↑1405.0 | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #32 / 104 | ↓18.05 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #14 / 241 | ↑51.94 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #156 / 241 | ↑86 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #49 / 241 | ↑89.44 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #36 / 239 | ↑99.09 | Source ↗official | |
| LiveSecBenchethics | #12 / 43 | ↑69.21 | Source ↗official | |
| LiveSecBenchfactuality | #13 / 43 | ↑62.29 | Source ↗official | |
| LiveSecBenchlegality | #12 / 43 | ↑67.96 | Source ↗official | |
| LiveSecBenchprivacy | #14 / 43 | ↑64.5 | Source ↗official | |
| LiveSecBenchpsychological_health | #10 / 43 | ↑69.47 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #54 / 113 | ↑6.82 | Source ↗official | |
| SpeciEvalland_animal_4ns | #100 / 113 | ↓4.97 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #105 / 113 | ↓5.2 | Source ↗official | |
| SpeciEvalspeciesism | #40 / 113 | ↓1.81 | 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.
UGI Political Values
Taiwan Sovereignty Benchmark Pro
| Dimension | Value | Distribution |
|---|---|---|
| Pro-Taiwan rubric compatibility | 55 | |
| Warning-phrase rate | 35 | |
| Soft-censorship rate | 0 | |
| API-error rate | 0 |