benchgap
Alibaba (Qwen) · model

Qwen3.6 Plus benchmark scores

As of 2026-10-07, Qwen3.6 Plus (Alibaba (Qwen)) has measured scores on 58 benchmarks and estimated scores on 78 more.

BenchmarkScoreSource
AA-GPQA Diamond88.2%measured
AA-HLE27.8%measured
AA-SciCode46.9%estimated ± 3.2 pp, high confidence
AIME2695.3%measured
Artificial Analysis Intelligence Index27.0%measured
CritPt2.9%measured
GDPval-AA24.7%measured
GPQA Diamond90.4%estimated ± 0.1 pp, high confidence
HLE28.8%measured
IFBench75.8%measured
AA-LCR78.3%measured
LiveCodeBench v687.1%measured
AA-Omniscience Accuracy26.4%measured
SciCode48.7%estimated ± 2.9 pp, medium confidence
Terminal-Bench 2.168.5%estimated ± 6.9 pp, medium confidence
BrowseComp66.9%estimated ± 4.4 pp, high confidence
HLE w/ tools45.0%estimated ± 2.4 pp, medium confidence
IFEval94.3%measured
LongBench v262.0%measured
VITA-Bench44.3%measured
τ²-bench results97.7%measured
AA Coding Index54.5%measured
AA-MMMU-Pro78.0%measured
APEX-Agents8.4%estimated ± 0.6 pp, low confidence
APEX-Agents-AA21.4%estimated ± 5.7 pp, medium confidence
BioMysteryBench (human-difficult)32.4%estimated ± 5.9 pp, low confidence
DeepSearchQA84.5%estimated ± 8.4 pp, medium confidence
SWE-bench Verified78.8%measured
AutomationBench28.4%estimated ± 9.7 pp, low confidence
BFCL v474.0%estimated ± 2.6 pp, high confidence
CyberGym51.7%estimated ± 6.2 pp, low confidence
JobBench18.5%estimated ± 5.1 pp, low confidence
SkillsBench46.4%estimated ± 5.6 pp, low confidence
SWE-bench Pro56.6%measured
τ³-bench results70.7%measured
WideResearch74.3%measured
AA Agentic Index18.7%estimated ± 2.8 pp, high confidence
MMLU-Pro88.5%measured
FrontierMath v2 (Tier 4)8.3%measured
FrontierMath v2 (Tiers 1-3)26.2%measured
GPQA90.4%measured
AA-IFBench75.2%measured
Gert Labs50.6%measured
AA-AnalystAgent17.2%estimated ± 12.0 pp, low confidence
AA EnterpriseOps-Gym51.4%estimated ± 3.6 pp, medium confidence
ApprenticeBench0.0%estimated ± 3.1 pp, low confidence
ARC-AGI-180.4%estimated ± 5.4 pp, medium confidence
ARC-AGI-255.6%estimated ± 11.6 pp, low confidence
CursorBench 3.147.9%estimated ± 4.8 pp, low confidence
CursorBench 3.253.1%estimated ± 3.6 pp, high confidence
Terminal-Bench 3.02.8%estimated ± 6.9 pp, low confidence
FrontierCode 1.1 Main21.1%estimated ± 2.7 pp, low confidence
FrontierSWE v29.1%estimated ± 9.0 pp, medium confidence
OfficeQA Pro50.5%estimated ± 7.2 pp, medium confidence
OSWorld-Verified60.9%estimated ± 5.9 pp, medium confidence
Vals GPQA Diamond87.4%measured
Vals LiveCodeBench86.0%measured
Vals MMLU-Pro87.7%measured
Vals SWE-bench73.4%measured
Terminal-Bench 2.1 (Vals AI)53.2%measured
VulcanBench v384.3%estimated ± 5.6 pp, low confidence
AA AutomationBench4.1%estimated ± 7.0 pp, medium confidence
GDP.pdf9.4%estimated ± 5.4 pp, medium confidence
AA ITBench22.7%estimated ± 3.6 pp, low confidence
AA Tau3 Banking22.0%estimated ± 5.8 pp, medium confidence
AA Terminal-Bench 2.165.8%estimated ± 4.4 pp, high confidence
AA Terminal-Bench 4.06.0%estimated ± 14.0 pp, low confidence
CursorBench 4.018.8%estimated ± 1.6 pp, medium confidence
CWE-bench v134.6%estimated ± 6.9 pp, low confidence
DeepSWE21.7%estimated ± 6.4 pp, low confidence
HLE w/o tools35.4%estimated ± 4.9 pp, high confidence
OSWorld 2.00.1%estimated ± 4.5 pp, low confidence
PostTrainBench v1.120.6%estimated ± 2.7 pp, high confidence
SWE Multilingual73.8%measured
Toolathlon-Verified49.1%estimated ± 2.4 pp, medium confidence
Vibe Code Bench25.6%measured
CharXiv81.5%measured
CharXiv w/o tools83.7%estimated ± 1.7 pp, medium confidence
C-Eval93.3%measured
Claw-Eval58.8%measured
DeepPlanning41.5%measured
HMMT Feb 202596.7%measured
HMMT Feb 202687.8%measured
HMMT Nov 202594.6%measured
MathVision88.0%measured
MCP Atlas48.2%measured
MCP-Tasks74.1%measured
MMAnswerBench83.8%measured
MMLU-ProX84.7%measured
MMLU-Redux94.5%measured
MMMU-Pro78.8%measured
NL2Repo45.2%estimated ± 3.1 pp, low confidence
NOVA-6357.9%measured
QwenClawBench57.2%measured
ScreenSpot Pro68.2%measured
SuperGPQA71.6%measured
Terminal-Bench 2.061.6%measured
Toolathlon39.8%measured
V*96.9%measured
VideoMMMU84.0%measured
ERQA66.0%estimated ± 2.4 pp, low confidence
HealthBench Hard4.2%estimated ± 5.1 pp, low confidence
LiveCodeBench Pro81.0%estimated ± 6.0 pp, low confidence
MedXpertQA (MM)72.0%estimated ± 2.4 pp, high confidence
MMLU-Pro (Arcee)88.0%estimated ± 4.0 pp, medium confidence
React Native Evals70.7%estimated ± 2.4 pp, low confidence
ResearchClawBench18.0%measured
SWE-Rebench61.7%estimated ± 3.6 pp, medium confidence
ARC-AGI-30.0%estimated ± 2.1 pp, medium confidence
FrontierMath (legacy)27.0%estimated ± 0.7 pp, low confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
DRACO71.7%estimated ± 3.0 pp, medium confidence
FrontierCode 1.1 Extended20.2%estimated ± 2.4 pp, low confidence
HealthBench (raw)36.9%estimated ± 6.4 pp, low confidence
HealthBench (length-adjusted)53.9%estimated ± 3.7 pp, low confidence
HealthBench Professional59.2%estimated ± 1.3 pp, medium confidence
HealthBench Professional (raw)48.1%estimated ± 5.0 pp, low confidence
HLE-Verified15.4%estimated ± 2.0 pp, low confidence
LABBench281.8%estimated ± 1.8 pp, low confidence
AIME 202595.9%estimated ± 1.5 pp, medium confidence
MMMU86.0%measured
LiveCodeBench100.0%estimated ± 8.9 pp, low confidence
OpenHarmony Bench49.5%estimated ± 2.8 pp, medium confidence
IMOAnswerBench82.4%estimated ± 5.9 pp, medium confidence
BabyVision54.0%estimated ± 4.3 pp, medium confidence
MMVU76.7%estimated ± 2.9 pp, medium confidence
SimpleVQA89.7%estimated ± 5.7 pp, low confidence
ZeroBench15.0%estimated ± 4.7 pp, low confidence
MMLU89.5%estimated ± 1.1 pp, medium confidence
MMMU-Pro w/ Python80.1%estimated ± 0.6 pp, high confidence
RefCOCO (avg)92.6%estimated ± 1.8 pp, low confidence
MATH-50097.7%estimated ± 3.1 pp, medium confidence
MathVision w/ Python94.0%estimated ± 0.5 pp, low confidence
RealWorldQA38.1%estimated ± 0.8 pp, low confidence
Video-MME (with subtitle)87.7%estimated ± 0.9 pp, medium confidence
OmniDocBench 1.591.7%estimated ± 5.9 pp, low confidence