benchgap
Xiaomi · model

MiMo-V2.5-Pro benchmark scores

As of 2026-11-22, MiMo-V2.5-Pro (Xiaomi) has measured scores on 29 benchmarks and estimated scores on 113 more.

BenchmarkScoreSource
AA-GPQA Diamond86.6%measured
AA-HLE35.7%measured
AA-SciCode50.6%measured
Artificial Analysis Intelligence Index26.0%measured
CritPt4.0%measured
GDPval-AA31.0%measured
GPQA Diamond86.6%estimated ± 1.8 pp, medium confidence
HLE48.0%measured
IFBench78.8%estimated ± 6.6 pp, medium confidence
AA-LCR79.7%measured
LiveCodeBench v687.1%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy22.4%measured
SciCode39.5%estimated ± 2.9 pp, low confidence
Terminal-Bench 2.174.8%estimated ± 6.7 pp, medium confidence
BrowseComp24.4%estimated ± 4.2 pp, medium confidence
HLE w/ tools26.8%estimated ± 4.1 pp, low confidence
IFEval93.2%estimated ± 2.8 pp, high confidence
LongBench v265.4%estimated ± 5.5 pp, medium confidence
VITA-Bench41.0%estimated ± 10.7 pp, low confidence
τ²-bench results94.2%measured
AA Coding Index60.2%measured
APEX-Agents23.3%estimated ± 9.4 pp, low confidence
APEX-Agents-AA2.4%measured
BioMysteryBench (human-difficult)34.6%estimated ± 6.8 pp, low confidence
DeepSearchQA85.9%estimated ± 4.5 pp, high confidence
SWE-bench Verified78.5%estimated ± 3.6 pp, high confidence
AutomationBench48.8%estimated ± 9.1 pp, low confidence
BFCL v474.5%estimated ± 2.7 pp, high confidence
CyberGym63.0%estimated ± 5.3 pp, low confidence
JobBench0.0%estimated ± 8.2 pp, low confidence
SWE-bench Pro57.2%measured
τ³-bench results72.9%measured
WideResearch78.4%estimated ± 3.3 pp, high confidence
AA Agentic Index22.7%measured
MMLU-Pro85.4%estimated ± 2.5 pp, medium confidence
AA-IFBench79.9%measured
FailureSensorIQ64.0%estimated ± 1.7 pp, high confidence
GPQA91.7%estimated ± 1.9 pp, high confidence
MMLU90.3%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond82.6%measured
Vals LiveCodeBench81.4%measured
Vals MMLU-Pro84.6%measured
MMMLU88.1%estimated ± 0.7 pp, medium confidence
Chess Puzzles19.4%estimated ± 12.3 pp, low confidence
AA MMLU-Pro84.7%estimated ± 0.7 pp, medium confidence
Gert Labs44.2%measured
AA-AnalystAgent44.8%estimated ± 5.9 pp, low confidence
AA Harvey LAB v1.082.3%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking29.0%estimated ± 3.7 pp, high confidence
AA Terminal-Bench 2.180.3%estimated ± 2.5 pp, medium confidence
ApprenticeBench5.2%estimated ± 10.4 pp, low confidence
ARC-AGI-185.1%estimated ± 3.9 pp, high confidence
ARC-AGI-249.7%estimated ± 13.0 pp, low confidence
CursorBench 3.145.5%estimated ± 3.4 pp, low confidence
CursorBench 3.254.1%estimated ± 3.6 pp, high confidence
Terminal-Bench 3.04.4%estimated ± 6.9 pp, low confidence
FrontierCode 1.1 Main28.1%estimated ± 4.6 pp, high confidence
FrontierSWE v210.7%estimated ± 8.3 pp, medium confidence
MazeBench0.0%measured
Mystery Game Puzzles13.6%estimated ± 14.6 pp, low confidence
NYT Connections34.4%measured
OSWorld-Verified71.0%estimated ± 3.8 pp, medium confidence
Terminal-Bench 4.12.2%estimated ± 9.2 pp, low confidence
Terminal-Bench Hard43.2%measured
Vals SWE-bench74.0%measured
Terminal-Bench 2.1 (Vals AI)57.3%measured
VulcanBench v381.6%estimated ± 5.6 pp, low confidence
CWE-Bench-AA49.1%estimated ± 8.1 pp, medium confidence
AA Cyber Index15.1%estimated ± 9.2 pp, low confidence
DeepsecBench-AA14.8%estimated ± 8.2 pp, medium confidence
HealthBench Professional (length-adjusted, Artificial Analysis)53.7%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)37.4%estimated ± 0.8 pp, low confidence
AA ITBench38.2%measured
Bug Hunt Bench8.6%estimated ± 9.0 pp, low confidence
CursorBench 4.033.5%estimated ± 3.3 pp, medium confidence
CWE-bench v157.4%estimated ± 5.1 pp, medium confidence
DeepSWE33.4%estimated ± 6.4 pp, low confidence
DRACO86.9%estimated ± 0.7 pp, low confidence
FrontierCode 2.020.4%estimated ± 6.7 pp, low confidence
HLE w/o tools34.0%measured
OSWorld 2.150.7%estimated ± 3.6 pp, medium confidence
OSWorld 2.1 Offline8.2%estimated ± 1.9 pp, low confidence
PostTrainBench v1.118.9%estimated ± 3.7 pp, medium confidence
StealthMark41.3%estimated ± 10.7 pp, low confidence
Toolathlon-Verified51.7%estimated ± 2.3 pp, low confidence
VulcanBench Frontier v480.8%estimated ± 6.8 pp, low confidence
AA AutomationBench51.0%estimated ± 9.9 pp, low confidence
GDP.pdf18.2%estimated ± 4.2 pp, high confidence
AA Terminal-Bench 4.03.7%estimated ± 13.7 pp, low confidence
ARC-AGI-31.2%estimated ± 7.2 pp, medium confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
CWE-bench43.2%estimated ± 6.2 pp, medium confidence
MKA34.3%estimated ± 4.6 pp, high confidence
SkillsBench official leaderboard (BenchFlow)42.5%estimated ± 6.1 pp, medium confidence
BinaryAudit12.4%estimated ± 12.1 pp, low confidence
OTelBench19.6%estimated ± 6.6 pp, medium confidence
Vibe Code Bench34.0%estimated ± 10.0 pp, low confidence
FACTS Multimodal21.1%estimated ± 4.0 pp, medium confidence
Hermes Bench60.6%estimated ± 2.9 pp, high confidence
Hermes Index37.2%estimated ± 3.1 pp, high confidence
ExploitGym8.3%estimated ± 2.7 pp, medium confidence
C-Eval93.3%estimated ± 1.0 pp, medium confidence
Claw-Eval63.8%measured
DeepPlanning35.4%estimated ± 8.3 pp, low confidence
MCP Atlas70.2%estimated ± 6.8 pp, medium confidence
MCP-Tasks79.7%estimated ± 3.6 pp, low confidence
MMLU-Redux93.8%estimated ± 0.8 pp, medium confidence
NL2Repo41.9%estimated ± 4.7 pp, high confidence
PostTrainBench v116.9%estimated ± 3.6 pp, high confidence
QwenClawBench60.8%estimated ± 6.0 pp, low confidence
SuperGPQA87.4%estimated ± 7.2 pp, medium confidence
Terminal-Bench 2.068.4%measured
Toolathlon49.4%estimated ± 5.1 pp, low confidence
AA LiveCodeBench88.4%estimated ± 2.6 pp, high confidence
HealthBench Hard21.7%estimated ± 8.6 pp, low confidence
LiveCodeBench Pro81.5%estimated ± 6.0 pp, low confidence
MedXpertQA (Text)51.0%estimated ± 8.9 pp, low confidence
MMLU-Pro (Arcee)92.1%estimated ± 4.0 pp, low confidence
React Native Evals79.2%estimated ± 3.2 pp, high confidence
ResearchClawBench17.4%estimated ± 1.3 pp, high confidence
SWE-Rebench58.1%estimated ± 8.9 pp, low confidence
PostTrainBench16.3%estimated ± 3.7 pp, medium confidence
Finance Agent v255.4%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
FrontierCode 1.1 Extended55.0%estimated ± 3.0 pp, low confidence
HealthBench (raw)44.6%estimated ± 6.0 pp, low confidence
HealthBench (length-adjusted)51.9%estimated ± 3.4 pp, low confidence
HealthBench Professional44.1%estimated ± 4.2 pp, low confidence
HealthBench Professional (raw)55.5%estimated ± 4.5 pp, medium confidence
HLE-Verified0.0%estimated ± 6.8 pp, low confidence
LABBench280.1%estimated ± 1.8 pp, low confidence
ScarfBench26.5%estimated ± 9.1 pp, low confidence
STEMBench2.3%estimated ± 0.2 pp, low confidence
SimpleQA21.4%estimated ± 4.3 pp, medium confidence
Agents' Last Exam30.5%estimated ± 9.7 pp, medium confidence
OpenHarmony Bench v1.056.4%estimated ± 2.4 pp, high confidence
MRCRv285.7%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.4%estimated ± 2.6 pp, low confidence
BioBasicsBench33.9%estimated ± 2.8 pp, low confidence
LifeVal75.2%estimated ± 3.8 pp, medium confidence
PinchBench94.6%estimated ± 5.5 pp, low confidence