Model profile
Evidence summary
Safety. MiMo v2.5 Pro has an estimated Safety rank of #71; its 90% source-sensitivity interval is #20–#219. Its behavior-only rank is #67; company governance moves the combined estimate to #71. Published Safety evidence spans 9 eval lineages and 6 of 7 components. Its strongest relative result is Concordia — SOSBench-Bio (safety_score, #1 of 63); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #253 of 272).
Freedom. MiMo v2.5 Pro has an estimated Freedom rank of #520; its 90% source-sensitivity interval is #303–#626. Published Freedom evidence spans 5 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #20 of 272); its weakest is Concordia — SOSBench-Bio (safety_score, #61 of 63).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗mimo-v2-5-pro
- Hugging Face ↗XiaomiMiMo/MiMo-V2.5-Pro
- OpenRouter ↗xiaomi/mimo-v2.5-pro
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
- mimo.xiaomi.com ↗
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 | #17 / 361 | ↓0.247 | Source ↗official | |
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #38 / 111 | ↑1445.0 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #36 / 122 | ↑0.5 | Source ↗official | |
| Concordia — Agentic-Misalignmentsafety_score | #30 / 54 | ↑75.67 | Source ↗official | |
| Concordia — AirBench-Deceptionsafety_score | #10 / 63 | ↑96.3 | Source ↗official | |
| Concordia — AirBench-Manipulationsafety_score | #8 / 56 | ↑98.67 | Source ↗official | |
| Concordia — AirBench-SecurityRiskssafety_score | #23 / 63 | ↑96.4 | Source ↗official | |
| Concordia — APEsafety_score | #22 / 55 | ↑39.3 | Source ↗official | |
| Concordia — CyberSecEval2-PromptInjectionsafety_score | #33 / 63 | ↑83.67 | Source ↗official | |
| Concordia — DarkBenchsafety_score | #4 / 55 | ↑69.09 | Source ↗official | |
| Concordia — Fortress-Biologicalsafety_score | #22 / 54 | ↑72.51 | Source ↗official | |
| Concordia — Fortress-Chemicalsafety_score | #16 / 54 | ↑74.9 | Source ↗official | |
| Concordia — Fortress-Privacy/Scamssafety_score | #11 / 54 | ↑76.57 | Source ↗official | |
| Concordia — FRT-AirBench-Manipulationsafety_score | #15 / 45 | ↑44.67 | Source ↗official | |
| Concordia — FRT-AirBench-SecurityRiskssafety_score | #22 / 45 | ↑29.83 | Source ↗official | |
| Concordia — FRT-SciKnowEval-BiologicalHarmfulQAsafety_score | #9 / 45 | ↑7.667 | Source ↗official | |
| Concordia — FRT-SOSBench-Chemsafety_score | #14 / 45 | ↑63.67 | Source ↗official | |
| Concordia — MASKsafety_score | #24 / 62 | ↑68.86 | Source ↗official | |
| Concordia — SciKnowEval-BiologicalHarmfulQAsafety_score | #27 / 63 | ↑84.85 | Source ↗official | |
| Concordia — SciKnowEval-ChemicalHarmfulQAsafety_score | #36 / 63 | ↑17.4 | Source ↗official | |
| Concordia — Shutdown-Resistancesafety_score | #41 / 53 | ↑88 | Source ↗official | |
| Concordia — SOSBench-Biosafety_score | #1 / 63 | ↑100 | Source ↗official | |
| Concordia — SOSBench-Chemsafety_score | #8 / 63 | ↑95.6 | Source ↗official | |
| DystopiaBenchbasaglia_score | #8 / 50 | ↓33.37 | Source ↗official | |
| DystopiaBenchbaudrillard_score | #20 / 50 | ↓51.07 | Source ↗official | |
| DystopiaBenchhuxley_score | #7 / 50 | ↓26.03 | Source ↗official | |
| DystopiaBenchlaguardia_score | #5 / 50 | ↓37.03 | Source ↗official | |
| DystopiaBenchorwell_score | #6 / 50 | ↓26.03 | Source ↗official | |
| DystopiaBenchpetrov_score | #10 / 50 | ↓42.53 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #60 / 272 | ↑28.94 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #253 / 272 | ↑53 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #33 / 272 | ↑97.78 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #63 / 270 | ↑98.18 | Source ↗official | |
| Opposite-Narrator Sycophancysycophancy_rate_pct | #11 / 24 | ↓2.6 | Source ↗official | |
| SM-Benchadversarial | #37 / 92 | ↑83.41 | Source ↗official | |
| SM-Benchambiguous_interpretation | #42 / 92 | ↑86.61 | Source ↗official | |
| SM-Benchanti_hallucination | #19 / 92 | ↑98.43 | Source ↗official | |
| SM-Bencheq_boundaries | #33 / 92 | ↑67.98 | Source ↗official | |
| SM-Benchoverfit | #26 / 92 | ↑85.79 | Source ↗official | |
| TACbase_welfare_rate | #68 / 94 | ↑23.7 | 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 |
|---|---|---|---|---|
| Concordia — AirBench-Deceptionsafety_score | #53 / 63 | ↓96.3 | Source ↗official | |
| Concordia — AirBench-Manipulationsafety_score | #46 / 56 | ↓98.67 | Source ↗official | |
| Concordia — AirBench-PoliticalPersuasionsafety_score | #37 / 56 | ↓70 | Source ↗official | |
| Concordia — AirBench-SecurityRiskssafety_score | #40 / 63 | ↓96.4 | Source ↗official | |
| Concordia — Fortress-Biologicalsafety_score | #33 / 54 | ↓72.51 | Source ↗official | |
| Concordia — Fortress-Chemicalsafety_score | #39 / 54 | ↓74.9 | Source ↗official | |
| Concordia — Fortress-Privacy/Scamssafety_score | #44 / 54 | ↓76.57 | Source ↗official | |
| Concordia — FRT-AirBench-Manipulationsafety_score | #31 / 45 | ↓44.67 | Source ↗official | |
| Concordia — FRT-AirBench-SecurityRiskssafety_score | #24 / 45 | ↓29.83 | Source ↗official | |
| Concordia — FRT-SciKnowEval-BiologicalHarmfulQAsafety_score | #37 / 45 | ↓7.667 | Source ↗official | |
| Concordia — FRT-SOSBench-Chemsafety_score | #32 / 45 | ↓63.67 | Source ↗official | |
| Concordia — SciKnowEval-BiologicalHarmfulQAsafety_score | #36 / 63 | ↓84.85 | Source ↗official | |
| Concordia — SciKnowEval-ChemicalHarmfulQAsafety_score | #28 / 63 | ↓17.4 | Source ↗official | |
| Concordia — SOSBench-Biosafety_score | #61 / 63 | ↓100 | Source ↗official | |
| Concordia — SOSBench-Chemsafety_score | #56 / 63 | ↓95.6 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #20 / 272 | ↓53 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #232 / 272 | ↓97.78 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #203 / 270 | ↓98.18 | Source ↗official | |
| SM-Benchadversarial | #53 / 92 | ↓83.41 | Source ↗official | |
| SM-Bencheq_boundaries | #33 / 92 | ↑67.98 | Source ↗official | |
| SM-Benchoverfit | #26 / 92 | ↑85.79 | Source ↗official | |
| SpeechMap model completioncomplete_pct | #164 / 181 | ↑32.4 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_adherence_score | #79 / 156 | ↑1.5 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_direct_score | #79 / 156 | ↑3 | 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.