MiMo-V2.6-Pro

Composite indexes

• Artificial Analysis Intelligence Index#15 of 151
  46independentv4.3.2Artificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#17 of 96
  34.8%independentAAArtificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Claude Sonnet 5.5 at 70.6%

• SciCode#6 of 100
  60.9%independentAAArtificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Claude Opus 5.5 at 66.9%

• Terminal-Bench 2.x#3 of 105
  89.9%vendorTerminal-Bench 2.1, vendor harnessHugging Face: XiaomiMiMo/MiMo-V2.6-Pro-RL ↗

  Leader: Claude Fable 5.1 at 91.4%

Agents and tool use

• GDPval-AA#8 of 109
  1673independentGDPval-AA v2.1Artificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Muse Wonder Full at 1857

• OSWorld-Verified#3 of 21
  82%vendorOSWorld-VerifiedHugging Face: XiaomiMiMo/MiMo-V2.6-Pro-RL ↗

  Leader: Kimi K3 at 84.8%

Reasoning and knowledge

• Humanity’s Last Exam#10 of 153
  49.4%independentAA, no toolsArtificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Claude Sonnet 5.5 at 64.5%

• CritPt#11 of 144
  26.6%independentAAArtificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Muse Wonder Full at 33%

Vision and long context

• AA Long Context Reasoning#3 of 144
  86.3%independentAAArtificial Analysis: MiMo-V2.6-Pro ↗

  Leader: Kimi K3 at 88.7%

Human preference

• LMArena WebDev#18 of 58
  1619independentarena entry "mimo-v2.6-pro", rank 19, 2156 votesLMArena WebDev leaderboard (fetched 2026-09-25) ↗

  Leader: Claude Opus 5.5 at 1818

Other published results