priorsBuilder

Xiaomi · released 2026-09-21

MiMo V2.6 Flash

Measured on 54 benchmarks.

General frontier

33.7

#72 of 136 models

includes estimates: 24% of the mix is measured for this model

Your mix’s benchmarks

2 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #116default of 58735.1%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #206default of 526-12.7

Leader: Opus 5.5 · 46.4

All 54 benchmarks

show 52 more

agentic

AA-Briefcase v1.1

224 results
  1. #36default of 2241493

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #32high of 500.4%

Leader: Opus 5.5 · 15.5%

agentic

AutomationBench-AA

219 results
  1. #19default of 21964.1%

Leader: Gemini 4 · 77.5%

agentic

BioMysteryBench

27 results
  1. #15high of 2769.3%

Leader: Sonnet 5.5 · 81.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #14high of 7911.3%

Leader: Muse Spark 1.2 · 25.4%

agentic

Legal Research Bench

78 results
  1. #37high of 7838.0%

Leader: Muse Spark 1.3 · 55.3%

agentic

Tax Agent Bench

69 results
  1. #29high of 6928.5%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #20high of 4724.2%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #30high of 404.3%

Leader: GPT-6 Astra · 62.9%

coding

Arena Image-to-WebDev

58 results
  1. #16high of 581597

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #15high of 1221637

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #29high of 7740.9%

Leader: Sonnet 5.5 · 69.8%

coding

IOI

44 results
  1. #27high of 4447.7%

Leader: Gemini 4 · 100.0%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #15high of 6217.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #27high of 620.5%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #19high of 6271.7%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #94default of 22651.3%

Leader: Opus 5.5 · 66.9%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #17high of 7476.4%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench v4-AA

216 results
  1. #55default of 21622.7%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #26high of 10579.0%

Leader: Sonnet 5.5 · 92.4%

composite

AA Intelligence Index v4.3.2

610 results
  1. #77default of 61037.9

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #66high of 2201451

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #74high of 2201452

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #20high of 4753.2%

Leader: Gemini 4 · 68.9%

cost

AA Intelligence Index Cost per Task

177 results
  1. #28default of 177$0.06

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #26default of 177$110

Leader: GPT-6 Luna · $10.63

economics

Excel Modeling Benchmark

74 results
  1. #23high of 7465.5%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #19high of 7856.3%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #134default of 2109.0%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #31default of 2881605

Leader: Opus 5.5 · 1866

knowledge

AA-Omniscience Accuracy

526 results
  1. #202default of 52627.0%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #138default of 52645.6%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

MedCode

94 results
  1. #54high of 9441.1%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #28high of 9685.3%

Leader: Opus 5.5 · 91.4%

knowledge

MMMU-Pro

265 results
  1. #123default of 26573.1%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #14high of 5067.6%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #160default of 52374.3%

Leader: Kimi K3 · 88.7%

math

ProofBench v1.1

52 results
  1. #22high of 5263.0%

Leader: Sonnet 5.5 · 100.0%

other

Arena Vision

116 results
  1. #52high of 1161248

Leader: Fable 5 · 1308

reasoning

CritPt

529 results
  1. #104default of 52912.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Humanity's Sixth Sense

25 results
  1. #18high of 2525.4%

Leader: GPT-6 Astra · 53.6%

reasoning

MysteryMechanism

27 results
  1. #20high of 2721.6%

Leader: GPT-6 Astra · 53.2%

reasoning

Roboflow Visual Reasoning

102 results
  1. #67high of 10258.5%
  2. #98low of 10233.1%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #47high of 8043.5%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #117high of 1712.7%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #166high of 18196.7%

Leader: GPT-5 Pro · 100.0%

safety

CyberBench v1.1

45 results
  1. #7high of 4575.4%

Leader: Wonder · 80.4%

safety

SRE Bench

32 results
  1. #16high of 324.2%

Leader: GPT-6 Astra · 56.9%

speed

Latency (Time to First Token)

351 results
  1. #208default of 3513.2s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #252default of 35162.8

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #130high of 1772.3s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #163high of 17628.5

Leader: Muse Spark 1.1 · 204.3