benchgap
Alibaba (Qwen) · model

Qwen3.8 Max Preview benchmark scores

As of 2026-11-22, Qwen3.8 Max Preview (Alibaba (Qwen)) has measured scores on 13 benchmarks and estimated scores on 110 more.

BenchmarkScoreSource
AA-GPQA Diamond92.8%measured
AA-HLE43.1%measured
AA-SciCode52.1%measured
Artificial Analysis Intelligence Index33.7%measured
CritPt17.7%measured
GDPval-AA58.3%measured
GPQA Diamond92.6%estimated ± 2.4 pp, high confidence
HLE49.1%estimated ± 8.5 pp, medium confidence
AA-LCR80.3%measured
LiveCodeBench v688.3%estimated ± 4.7 pp, high confidence
AA-Omniscience Accuracy29.5%measured
SciCode51.9%estimated ± 3.1 pp, high confidence
BrowseComp88.8%estimated ± 4.2 pp, medium confidence
HLE w/ tools57.0%estimated ± 4.1 pp, low confidence
LongBench v265.9%estimated ± 5.5 pp, medium confidence
τ²-bench results95.5%estimated ± 7.0 pp, medium confidence
AA Coding Index71.8%measured
AA-MMMU-Pro82.8%measured
APEX-Agents39.1%estimated ± 4.9 pp, low confidence
APEX-Agents-AA42.4%measured
BioMysteryBench (human-difficult)47.5%estimated ± 6.8 pp, low confidence
DeepSearchQA93.3%estimated ± 4.5 pp, high confidence
SWE-bench Verified83.4%estimated ± 6.3 pp, medium confidence
BFCL v477.2%estimated ± 2.7 pp, medium confidence
CyberGym83.6%estimated ± 5.3 pp, medium confidence
JobBench55.9%estimated ± 8.2 pp, low confidence
SWE-bench Pro64.4%estimated ± 6.0 pp, medium confidence
WideResearch84.8%estimated ± 7.8 pp, low confidence
AA Agentic Index49.6%measured
MMLU-Pro85.3%estimated ± 5.0 pp, medium confidence
GPQA93.0%estimated ± 6.1 pp, medium confidence
MMLU90.5%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond91.9%estimated ± 3.2 pp, high confidence
Vals LiveCodeBench84.6%estimated ± 5.7 pp, low confidence
Vals MMLU-Pro84.7%estimated ± 2.9 pp, high confidence
MMMLU90.6%estimated ± 0.7 pp, low confidence
MMMU88.1%estimated ± 2.7 pp, medium confidence
AA MMLU-Pro85.5%estimated ± 0.8 pp, high confidence
Gert Labs59.8%estimated ± 9.3 pp, medium confidence
AA Harvey LAB v1.093.4%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking46.3%estimated ± 2.7 pp, low confidence
ApprenticeBench20.3%estimated ± 5.0 pp, low confidence
ARC-AGI-188.0%estimated ± 7.8 pp, low confidence
ARC-AGI-264.6%estimated ± 14.4 pp, low confidence
CursorBench 3.149.2%estimated ± 3.4 pp, medium confidence
CursorBench 3.259.7%estimated ± 4.4 pp, high confidence
FrontierCode 1.1 Main41.6%estimated ± 4.6 pp, high confidence
FrontierSWE v222.6%estimated ± 14.2 pp, low confidence
MazeBench7.9%estimated ± 5.5 pp, low confidence
OfficeQA Pro67.5%estimated ± 6.8 pp, medium confidence
OSWorld-Verified86.3%estimated ± 8.4 pp, low confidence
ParseBench-169.0%estimated ± 6.2 pp, medium confidence
Vals SWE-bench85.9%estimated ± 5.5 pp, medium confidence
CWE-Bench-AA58.7%estimated ± 10.1 pp, low confidence
AA Cyber Index29.0%estimated ± 9.2 pp, low confidence
DeepsecBench-AA27.4%estimated ± 8.9 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)56.1%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)63.9%estimated ± 0.8 pp, low confidence
AA ITBench48.7%estimated ± 4.0 pp, high confidence
Bug Hunt Bench21.8%estimated ± 9.0 pp, low confidence
Chartography (tools)85.1%estimated ± 4.1 pp, low confidence
CursorBench 4.034.4%estimated ± 2.0 pp, medium confidence
CWE-bench v161.1%estimated ± 6.4 pp, medium confidence
DeepSWE59.3%estimated ± 6.4 pp, medium confidence
DRACO86.0%estimated ± 2.0 pp, medium confidence
FrontierCode 2.026.1%estimated ± 9.3 pp, low confidence
HLE w/o tools42.8%estimated ± 4.9 pp, high confidence
OSWorld 2.172.8%estimated ± 6.0 pp, medium confidence
OSWorld 2.1 Offline76.7%estimated ± 1.9 pp, medium confidence
PostTrainBench v1.127.0%estimated ± 5.7 pp, medium confidence
Toolathlon-Verified74.0%estimated ± 2.3 pp, medium confidence
VulcanBench Frontier v483.6%estimated ± 4.0 pp, low confidence
GDP.pdf17.1%estimated ± 4.6 pp, high confidence
ARC-AGI-31.3%estimated ± 9.8 pp, medium confidence
ARC-AGI-3 Provider Adapter3.0%estimated ± 7.3 pp, medium confidence
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
MKA43.3%estimated ± 6.4 pp, low confidence
Vibe Code Bench59.1%estimated ± 10.0 pp, low confidence
Hermes Bench72.8%estimated ± 2.9 pp, high confidence
Hermes Index46.3%estimated ± 8.7 pp, medium confidence
ParseBench-276.0%estimated ± 6.2 pp, medium confidence
CharXiv85.7%estimated ± 6.3 pp, low confidence
CharXiv w/o tools85.0%estimated ± 2.5 pp, high confidence
ExploitGym25.6%estimated ± 10.6 pp, low confidence
AI2D_TEST94.2%estimated ± 2.0 pp, low confidence
C-Eval95.6%estimated ± 1.0 pp, low confidence
CountBench99.9%estimated ± 2.9 pp, low confidence
DynaMath85.5%estimated ± 3.0 pp, low confidence
ERQA68.8%estimated ± 3.1 pp, medium confidence
MathVision92.9%estimated ± 3.1 pp, medium confidence
MCP Atlas78.7%estimated ± 7.8 pp, low confidence
MMLU-Redux94.7%estimated ± 0.8 pp, medium confidence
MMMU-Pro82.9%estimated ± 1.5 pp, high confidence
NL2Repo53.1%estimated ± 4.7 pp, high confidence
RealWorldQA87.9%estimated ± 2.9 pp, low confidence
ScreenSpot Pro79.1%estimated ± 10.8 pp, low confidence
SuperGPQA68.0%estimated ± 9.3 pp, medium confidence
Toolathlon55.9%estimated ± 6.3 pp, low confidence
V*100.0%estimated ± 7.5 pp, low confidence
Video-MME (w/o subtitle)88.8%estimated ± 1.7 pp, low confidence
Video-MME (with subtitle)89.9%estimated ± 3.7 pp, low confidence
HealthBench Hard32.0%estimated ± 8.6 pp, low confidence
MedXpertQA (MM)77.6%estimated ± 6.5 pp, low confidence
React Native Evals82.2%estimated ± 3.2 pp, high confidence
ResearchClawBench21.1%estimated ± 1.9 pp, medium confidence
PostTrainBench25.0%estimated ± 5.9 pp, low confidence
BioMysteryBench (human-solvable)88.7%estimated ± 0.6 pp, medium confidence
FrontierCode 1.1 Extended57.7%estimated ± 3.4 pp, low confidence
HealthBench (raw)51.5%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)54.8%estimated ± 3.4 pp, medium confidence
HealthBench Professional57.1%estimated ± 4.2 pp, medium confidence
HealthBench Professional (raw)62.0%estimated ± 4.5 pp, high confidence
HLE-Verified52.7%estimated ± 2.0 pp, medium confidence
LABBench281.5%estimated ± 1.8 pp, low confidence
ScarfBench46.5%estimated ± 9.1 pp, low confidence
STEMBench2.4%estimated ± 0.2 pp, medium confidence
SimpleQA31.2%estimated ± 4.3 pp, medium confidence
BabyVision w/ Python92.5%estimated ± 4.0 pp, low confidence
OpenHarmony Bench v1.056.0%measured
MRCRv285.8%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.4%estimated ± 2.6 pp, low confidence
BioBasicsBench34.1%estimated ± 2.8 pp, medium confidence
MMMU-Pro w/ Python85.2%estimated ± 2.0 pp, high confidence