benchgap
Xiaomi · model

MiMo-V2.6-Pro benchmark scores

As of 2026-11-21, MiMo-V2.6-Pro (Xiaomi) has measured scores on 34 benchmarks and estimated scores on 104 more.

BenchmarkScoreSource
AA-GPQA Diamond91.8%estimated ± 9.2 pp, medium confidence
AA-HLE49.4%measured
AA-SciCode60.9%measured
Artificial Analysis Intelligence Index46.3%measured
CritPt26.6%measured
GDPval-AA59.2%measured
GPQA Diamond94.0%estimated ± 2.4 pp, high confidence
HLE57.5%estimated ± 8.5 pp, medium confidence
AA-LCR86.3%measured
LiveCodeBench v6100.0%estimated ± 4.9 pp, medium confidence
AA-Omniscience Accuracy34.8%measured
SciCode60.4%estimated ± 3.1 pp, medium confidence
Terminal-Bench 2.189.9%measured
BrowseComp86.6%estimated ± 3.0 pp, medium confidence
HLE w/ tools63.8%estimated ± 4.9 pp, low confidence
LongBench v271.5%estimated ± 5.5 pp, low confidence
AA Coding Index76.7%estimated ± 2.5 pp, high confidence
APEX-Agents68.4%estimated ± 7.4 pp, low confidence
APEX-Agents-AA41.7%estimated ± 3.6 pp, low confidence
BioMysteryBench (human-difficult)49.3%estimated ± 6.8 pp, low confidence
DeepSearchQA87.7%estimated ± 7.9 pp, low confidence
SWE-bench Verified89.0%estimated ± 5.1 pp, medium confidence
AutomationBench53.1%measured
BFCL v477.2%estimated ± 2.7 pp, medium confidence
CyberGym94.0%measured
JobBench62.0%measured
SWE-bench Pro77.5%estimated ± 6.3 pp, medium confidence
WideResearch83.9%estimated ± 1.2 pp, low confidence
AA Agentic Index56.4%estimated ± 3.1 pp, high confidence
MMLU-Pro85.5%estimated ± 5.0 pp, medium confidence
GPQA92.8%estimated ± 7.8 pp, medium confidence
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond92.1%estimated ± 5.7 pp, medium confidence
Vals LiveCodeBench89.0%estimated ± 5.7 pp, low confidence
Vals MMLU-Pro86.3%estimated ± 2.9 pp, high confidence
MMMLU91.4%estimated ± 0.7 pp, low confidence
Chess Puzzles41.1%estimated ± 8.0 pp, medium confidence
AA MMLU-Pro86.4%estimated ± 0.8 pp, high confidence
Gert Labs73.1%estimated ± 5.5 pp, low confidence
AA-AnalystAgent37.5%measured
AA EnterpriseOps-Gym46.7%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking100.0%estimated ± 2.4 pp, low confidence
AA Terminal-Bench 2.182.1%estimated ± 2.5 pp, high confidence
ApprenticeBench35.4%estimated ± 4.0 pp, medium confidence
ARC-AGI-194.7%estimated ± 7.8 pp, low confidence
ARC-AGI-281.7%estimated ± 14.4 pp, low confidence
CursorBench 3.170.9%estimated ± 3.4 pp, medium confidence
CursorBench 3.227.5%estimated ± 3.6 pp, low confidence
EBR-bench47.1%estimated ± 9.9 pp, medium confidence
Terminal-Bench 3.024.8%estimated ± 3.2 pp, high confidence
FrontierCode 1.1 Main50.8%estimated ± 3.5 pp, low confidence
FrontierSWE v231.0%estimated ± 11.1 pp, low confidence
Furniture Assembly46.8%estimated ± 6.5 pp, medium confidence
MaxxBench55.2%estimated ± 5.7 pp, medium confidence
MazeBench2.0%estimated ± 3.7 pp, high confidence
MirrorCode34.5%estimated ± 14.9 pp, low confidence
Mystery Game Puzzles27.0%estimated ± 7.5 pp, medium confidence
NYT Connections100.0%estimated ± 8.9 pp, low confidence
OSWorld-Verified82.0%measured
Terminal-Bench 4.127.0%measured
Terminal-Bench Hard59.9%estimated ± 2.8 pp, medium confidence
Vals SWE-bench94.2%estimated ± 8.2 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)67.8%measured
VulcanBench v378.2%estimated ± 3.9 pp, medium confidence
CWE-Bench-AA63.3%measured
AA Cyber Index56.1%measured
DeepsecBench-AA26.2%measured
Harvey LAB-AA v1.10.8%estimated ± 3.5 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)60.0%estimated ± 4.3 pp, medium confidence
HealthBench Professional (raw, Artificial Analysis)68.7%estimated ± 0.8 pp, medium confidence
AA ITBench44.3%estimated ± 4.7 pp, low confidence
Bug Hunt Bench22.7%measured
CursorBench 4.043.3%estimated ± 4.7 pp, high confidence
CWE-bench v155.0%estimated ± 4.5 pp, medium confidence
DeepSWE71.9%measured
DRACO86.2%estimated ± 2.0 pp, medium confidence
FrontierCode 2.023.8%estimated ± 6.7 pp, medium confidence
FrontierPhysics1.2%measured
HLE w/o tools53.0%estimated ± 4.9 pp, high confidence
OSWorld 2.016.7%estimated ± 3.6 pp, high confidence
OSWorld 2.173.5%estimated ± 6.0 pp, medium confidence
OSWorld 2.1 Offline78.0%estimated ± 1.9 pp, medium confidence
PostTrainBench v1.139.9%estimated ± 4.5 pp, high confidence
ProgramBench26.5%measured
SpireClimb9.3%estimated ± 5.4 pp, low confidence
StealthMark47.9%measured
Terminal-Bench 4.034.9%measured
Terminal-Bench-Science 0.14.3%estimated ± 10.7 pp, low confidence
Toolathlon-Verified76.9%measured
VulcanBench Frontier v488.5%estimated ± 1.2 pp, low confidence
AA AutomationBench58.6%measured
GDP.pdf19.2%measured
AA Terminal-Bench 4.034.8%measured
ARC-AGI-37.5%estimated ± 9.8 pp, medium confidence
ARC-AGI-3 Provider Adapter62.3%estimated ± 7.3 pp, medium confidence
cursorBench4152.9%estimated ± 3.7 pp, high confidence
CWE-bench48.1%estimated ± 4.4 pp, high confidence
MKA37.3%measured
FACTS Multimodal40.0%estimated ± 2.7 pp, medium confidence
Hermes Bench72.7%estimated ± 1.5 pp, medium confidence
Hermes Index37.4%estimated ± 2.7 pp, medium confidence
ExploitBench47.9%measured
ExploitGym17.8%measured
C-Eval93.3%estimated ± 1.5 pp, low confidence
Claw-Eval77.1%estimated ± 4.5 pp, medium confidence
MCP Atlas83.8%estimated ± 7.5 pp, medium confidence
MMLU-Redux95.3%estimated ± 0.8 pp, medium confidence
NL2Repo61.6%estimated ± 6.0 pp, medium confidence
SuperGPQA73.2%estimated ± 9.3 pp, medium confidence
Terminal-Bench 2.071.6%estimated ± 5.8 pp, medium confidence
Toolathlon55.9%estimated ± 6.3 pp, low confidence
HealthBench Hard33.6%estimated ± 8.6 pp, low confidence
React Native Evals81.3%estimated ± 4.7 pp, low confidence
ResearchClawBench19.7%estimated ± 2.1 pp, medium confidence
PostTrainBench33.1%estimated ± 5.0 pp, medium confidence
DAYJOB1.8%estimated ± 1.1 pp, medium confidence
DAYJOB Healthcare1.5%estimated ± 1.5 pp, medium confidence
BioMysteryBench (human-solvable)89.5%estimated ± 0.6 pp, medium confidence
HealthBench (raw)56.9%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)57.2%estimated ± 3.4 pp, medium confidence
HealthBench Professional59.4%estimated ± 4.2 pp, medium confidence
HealthBench Professional (raw)67.1%estimated ± 4.5 pp, high confidence
HLE-Verified54.4%estimated ± 6.8 pp, low confidence
LABBench283.8%estimated ± 2.4 pp, medium confidence
ScarfBench67.5%estimated ± 3.8 pp, medium confidence
STEMBench2.2%measured
SimpleQA38.4%estimated ± 4.3 pp, low confidence
Agents' Last Exam31.6%measured
OpenHarmony Bench v1.058.9%estimated ± 2.4 pp, medium confidence
OpenHarmony Bench v1.263.8%estimated ± 11.0 pp, low confidence
SEC-Bench Pro66.3%measured
MRCRv286.0%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.7%estimated ± 2.6 pp, low confidence
SWE-Marathon44.5%estimated ± 2.1 pp, medium confidence
BioBasicsBench44.9%estimated ± 2.8 pp, medium confidence
LifeVal71.5%estimated ± 2.3 pp, low confidence
ExploitBench Internal Port3.6%estimated ± 6.5 pp, medium confidence
SWE-Atlas Codebase QnA61.0%estimated ± 4.8 pp, high confidence