benchgap
Alibaba (Qwen) · model

Qwen3.8-Flash-Next benchmark scores

As of 2026-11-21, Qwen3.8-Flash-Next (Alibaba (Qwen)) has measured scores on 33 benchmarks and estimated scores on 123 more.

BenchmarkScoreSource
AA-GPQA Diamond92.3%measured
AA-HLE38.0%measured
AA-SciCode50.6%measured
Artificial Analysis Intelligence Index28.6%measured
CritPt11.1%measured
GDPval-AA56.4%measured
GPQA Diamond91.7%measured
HLE35.9%measured
IFBench81.3%measured
AA-LCR79.7%measured
LiveCodeBench v691.9%measured
AA-Omniscience Accuracy23.5%measured
SciCode50.4%estimated ± 3.1 pp, high confidence
BrowseComp82.9%estimated ± 1.9 pp, high confidence
HLE w/ tools61.1%estimated ± 5.3 pp, medium confidence
IFEval94.8%estimated ± 3.6 pp, high confidence
LongBench v265.4%estimated ± 5.5 pp, medium confidence
AA Coding Index73.1%measured
AA-MMMU-Pro78.6%measured
APEX-Agents50.9%estimated ± 9.4 pp, low confidence
APEX-Agents-AA36.3%estimated ± 2.9 pp, medium confidence
BioMysteryBench (human-difficult)44.6%estimated ± 6.8 pp, low confidence
DeepSearchQA87.7%estimated ± 8.2 pp, low confidence
SWE-bench Verified80.6%estimated ± 2.3 pp, high confidence
AutomationBench36.1%estimated ± 7.1 pp, low confidence
BFCL v477.1%estimated ± 2.7 pp, medium confidence
CyberGym83.4%estimated ± 3.1 pp, low confidence
JobBench55.7%measured
SWE-bench Pro62.5%measured
WideResearch82.5%estimated ± 1.2 pp, medium confidence
AA Agentic Index52.4%estimated ± 1.4 pp, high confidence
MMLU-Pro85.5%estimated ± 2.5 pp, medium confidence
AA-IFBench80.4%estimated ± 7.4 pp, medium confidence
GPQA91.7%measured
MMLU90.4%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond90.8%estimated ± 2.1 pp, high confidence
Vals LiveCodeBench86.1%estimated ± 3.4 pp, low confidence
Vals MMLU-Pro86.8%estimated ± 0.9 pp, high confidence
MMMLU89.7%estimated ± 0.7 pp, medium confidence
MMMU85.6%estimated ± 2.7 pp, high confidence
AA MMLU-Pro84.6%estimated ± 0.8 pp, high confidence
Gert Labs70.9%estimated ± 4.6 pp, medium confidence
AA-AnalystAgent41.7%estimated ± 8.0 pp, low confidence
AA EnterpriseOps-Gym46.4%estimated ± 1.2 pp, medium confidence
AA Harvey LAB v1.089.7%estimated ± 1.3 pp, medium confidence
AA Tau3 Banking45.4%measured
AA Terminal-Bench 2.186.1%measured
ApprenticeBench17.7%estimated ± 6.0 pp, low confidence
ARC-AGI-191.8%estimated ± 2.5 pp, high confidence
ARC-AGI-266.3%estimated ± 9.7 pp, medium confidence
CursorBench 3.145.5%estimated ± 3.4 pp, low confidence
CursorBench 3.261.1%estimated ± 3.6 pp, high confidence
Terminal-Bench 3.024.5%estimated ± 8.6 pp, low confidence
FrontierCode 1.1 Main43.5%estimated ± 3.5 pp, low confidence
FrontierSWE v219.9%estimated ± 8.3 pp, medium confidence
MazeBench2.0%estimated ± 4.0 pp, high confidence
OfficeQA Pro50.8%estimated ± 6.8 pp, medium confidence
OSWorld-Verified82.8%estimated ± 3.0 pp, high confidence
ParseBench-170.6%estimated ± 1.0 pp, medium confidence
Vals SWE-bench81.8%estimated ± 5.2 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)76.9%estimated ± 5.5 pp, low confidence
CWE-Bench-AA46.5%estimated ± 4.5 pp, medium confidence
AA Cyber Index7.8%estimated ± 13.3 pp, low confidence
DeepsecBench-AA18.6%estimated ± 5.2 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)54.5%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)56.9%estimated ± 0.8 pp, low confidence
AA ITBench47.8%estimated ± 5.1 pp, low confidence
Bug Hunt Bench14.9%estimated ± 7.2 pp, low confidence
Chartography (tools)82.5%estimated ± 4.1 pp, low confidence
CursorBench 4.035.2%estimated ± 2.0 pp, medium confidence
CWE-bench v159.9%estimated ± 6.4 pp, medium confidence
DeepSWE58.7%measured
DRACO85.6%estimated ± 2.0 pp, medium confidence
FrontierCode 2.022.1%estimated ± 6.7 pp, medium confidence
HLE w/o tools35.9%measured
OSWorld 2.019.4%measured
OSWorld 2.171.3%estimated ± 6.0 pp, medium confidence
OSWorld 2.1 Offline73.5%estimated ± 1.9 pp, medium confidence
PostTrainBench v1.127.2%estimated ± 4.4 pp, high confidence
SWE Multilingual81.0%measured
Toolathlon-Verified73.5%measured
VulcanBench Frontier v484.5%estimated ± 4.0 pp, low confidence
GDP.pdf17.0%estimated ± 4.1 pp, medium confidence
ARC-AGI-31.7%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-bench44.7%estimated ± 4.6 pp, medium confidence
MKA42.2%estimated ± 6.4 pp, low confidence
Vibe Code Bench63.2%estimated ± 10.0 pp, low confidence
Hermes Bench71.6%estimated ± 1.5 pp, medium confidence
Hermes Index42.4%estimated ± 7.0 pp, low confidence
ParseBench-269.7%estimated ± 6.2 pp, medium confidence
CharXiv90.6%measured
CharXiv w/o tools84.6%measured
ExploitGym11.0%estimated ± 10.4 pp, low confidence
AI2D_TEST93.5%estimated ± 0.9 pp, low confidence
C-Eval94.5%estimated ± 0.9 pp, low confidence
Claw-Eval74.0%estimated ± 4.5 pp, medium confidence
CountBench97.3%estimated ± 1.3 pp, low confidence
DynaMath88.1%estimated ± 0.8 pp, low confidence
ERQA72.3%measured
INCLUDE86.7%estimated ± 2.0 pp, medium confidence
MathVision90.6%measured
MCP Atlas80.9%estimated ± 6.0 pp, low confidence
MLVU (M-Avg)87.6%estimated ± 1.5 pp, high confidence
MMLU-ProX87.1%estimated ± 1.3 pp, medium confidence
MMLU-Redux93.9%estimated ± 0.8 pp, medium confidence
MMMU-Pro79.4%estimated ± 1.5 pp, high confidence
NL2Repo48.1%measured
OmniDocBench 1.591.7%estimated ± 1.8 pp, low confidence
PostTrainBench v127.8%estimated ± 4.6 pp, high confidence
RealWorldQA88.5%measured
ScreenSpot Pro85.1%estimated ± 4.8 pp, low confidence
SimpleVQA74.1%estimated ± 7.3 pp, medium confidence
SuperGPQA76.0%estimated ± 7.2 pp, medium confidence
Toolathlon55.9%estimated ± 6.3 pp, low confidence
V*100.0%estimated ± 7.5 pp, medium confidence
Video-MME (w/o subtitle)85.6%estimated ± 1.7 pp, medium confidence
Video-MME (with subtitle)88.9%estimated ± 1.9 pp, high confidence
VideoMMMU87.4%estimated ± 1.5 pp, medium confidence
We-Math87.5%estimated ± 2.1 pp, low confidence
HealthBench Hard23.3%estimated ± 5.1 pp, low confidence
LiveCodeBench Pro85.6%estimated ± 6.0 pp, low confidence
MedXpertQA (MM)78.5%estimated ± 4.2 pp, medium confidence
MedXpertQA (Text)59.2%estimated ± 4.5 pp, medium confidence
MMLU-Pro (Arcee)87.4%estimated ± 3.4 pp, low confidence
React Native Evals82.5%estimated ± 3.2 pp, high confidence
ResearchClawBench20.7%estimated ± 1.9 pp, medium confidence
PostTrainBench25.1%estimated ± 3.7 pp, high confidence
BioMysteryBench (human-solvable)84.3%estimated ± 0.6 pp, low confidence
FrontierCode 1.1 Extended56.8%estimated ± 3.0 pp, low confidence
HealthBench (raw)46.8%estimated ± 6.0 pp, low confidence
HealthBench (length-adjusted)52.8%estimated ± 3.4 pp, low confidence
HealthBench Professional45.9%estimated ± 4.2 pp, low confidence
HealthBench Professional (raw)57.6%estimated ± 4.5 pp, medium confidence
HLE-Verified49.9%estimated ± 2.0 pp, medium confidence
LABBench280.7%estimated ± 1.8 pp, low confidence
ScarfBench62.8%estimated ± 3.8 pp, medium confidence
STEMBench2.4%estimated ± 0.2 pp, low confidence
SimpleQA22.9%estimated ± 4.3 pp, medium confidence
Agents' Last Exam51.2%measured
BabyVision w/ Python89.7%estimated ± 3.7 pp, low confidence
OpenHarmony Bench v1.056.4%estimated ± 2.4 pp, high confidence
BabyVision62.8%estimated ± 3.7 pp, high confidence
MMVU79.5%estimated ± 3.3 pp, medium confidence
PerceptionBench58.0%estimated ± 3.6 pp, medium confidence
ZeroBench15.7%estimated ± 3.6 pp, medium confidence
MRCRv285.7%estimated ± 11.3 pp, low confidence
LVBench76.6%measured
MMLongBench-Doc61.4%estimated ± 2.6 pp, low confidence
RefCOCO (avg)91.3%estimated ± 3.1 pp, low confidence
OCRBench V275.4%estimated ± 5.4 pp, low confidence
BioBasicsBench27.2%estimated ± 2.2 pp, low confidence
MMMU-Pro w/ Python81.5%estimated ± 2.0 pp, high confidence
MathVision w/ Python95.7%measured
SWE-Atlas Codebase QnA55.4%estimated ± 4.8 pp, high confidence