benchgap
Alibaba (Qwen) · model

Qwen3.5 Flash benchmark scores

As of 2026-11-22, Qwen3.5 Flash (Alibaba (Qwen)) has measured scores on 7 benchmarks and estimated scores on 62 more.

BenchmarkScoreSource
AA-GPQA Diamond83.6%estimated ± 3.6 pp, high confidence
AA-HLE22.4%estimated ± 6.2 pp, medium confidence
AA-SciCode45.2%estimated ± 4.2 pp, high confidence
AIME2691.7%estimated ± 1.4 pp, medium confidence
Artificial Analysis Intelligence Index20.9%estimated ± 5.8 pp, medium confidence
CritPt2.5%estimated ± 5.3 pp, medium confidence
GPQA Diamond86.7%estimated ± 1.8 pp, medium confidence
HLE35.2%estimated ± 9.2 pp, low confidence
LiveCodeBench v688.0%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy31.5%estimated ± 6.5 pp, medium confidence
SciCode43.3%estimated ± 2.9 pp, low confidence
AA Coding Index44.1%estimated ± 5.6 pp, medium confidence
SWE-bench Verified73.9%estimated ± 4.4 pp, high confidence
SWE-bench Pro57.2%estimated ± 4.7 pp, high confidence
MMLU-Pro84.4%estimated ± 4.8 pp, high confidence
FailureSensorIQ64.3%estimated ± 2.1 pp, high confidence
GPQA84.8%estimated ± 4.5 pp, high confidence
MMLU92.0%estimated ± 1.2 pp, medium confidence
Vals GPQA Diamond82.8%measured
Vals LiveCodeBench83.3%measured
Vals MMLU-Pro84.1%measured
Chess Puzzles14.0%measured
AA MMLU-Pro86.2%estimated ± 1.1 pp, high confidence
ARC-AGI-173.6%estimated ± 6.5 pp, medium confidence
ARC-AGI-227.7%estimated ± 13.8 pp, low confidence
CursorBench 3.138.1%estimated ± 5.8 pp, low confidence
CursorBench 3.249.6%estimated ± 5.4 pp, low confidence
EBR-bench8.9%estimated ± 10.4 pp, low confidence
FrontierCode 1.1 Main16.6%estimated ± 5.2 pp, low confidence
FrontierMath v2 (Tier 4)0.0%estimated ± 4.0 pp, medium confidence
FrontierMath v2 (Tiers 1-3)18.2%measured
FrontierSWE v25.0%estimated ± 9.5 pp, low confidence
Furniture Assembly33.0%estimated ± 9.3 pp, medium confidence
MaxxBench47.3%estimated ± 8.8 pp, medium confidence
Mystery Game Puzzles20.0%measured
NYT Connections51.6%estimated ± 12.6 pp, low confidence
Vals SWE-bench64.4%measured
VulcanBench v382.6%estimated ± 5.6 pp, low confidence
CursorBench 4.030.7%estimated ± 3.4 pp, medium confidence
DeepSWE56.3%estimated ± 7.2 pp, low confidence
FrontierPhysics2.5%estimated ± 5.0 pp, medium confidence
HLE w/o tools28.3%estimated ± 4.9 pp, high confidence
PostTrainBench v1.110.8%estimated ± 3.7 pp, medium confidence
SpireClimb0.0%estimated ± 9.3 pp, medium confidence
StealthMark38.9%estimated ± 9.2 pp, medium confidence
ARC-AGI-30.0%estimated ± 8.6 pp, low confidence
SkillsBench official leaderboard (BenchFlow)43.8%estimated ± 6.2 pp, medium confidence
CompileBench84.4%estimated ± 10.0 pp, low confidence
OTelBench22.6%estimated ± 4.1 pp, medium confidence
FACTS Multimodal34.0%estimated ± 5.6 pp, low confidence
ExploitBench36.7%estimated ± 7.2 pp, low confidence
HMMT Feb 202681.9%estimated ± 3.1 pp, medium confidence
HMMT Nov 202588.3%estimated ± 1.5 pp, low confidence
MMAnswerBench78.4%estimated ± 2.4 pp, low confidence
NL2Repo38.2%estimated ± 4.7 pp, medium confidence
PostTrainBench v15.1%estimated ± 3.6 pp, medium confidence
AA LiveCodeBench84.7%estimated ± 2.6 pp, high confidence
React Native Evals74.1%estimated ± 4.7 pp, medium confidence
PostTrainBench8.7%estimated ± 3.7 pp, medium confidence
DAYJOB1.3%estimated ± 3.2 pp, high confidence
DAYJOB Healthcare0.7%estimated ± 4.2 pp, high confidence
HLE-Verified0.0%estimated ± 4.6 pp, low confidence
LABBench279.8%estimated ± 1.8 pp, low confidence
ScarfBench36.3%estimated ± 12.3 pp, low confidence
BlenderWorld (text-only)70.1%estimated ± 2.8 pp, medium confidence
Cookie-Bench77.1%estimated ± 1.3 pp, medium confidence
OpenHarmony Bench v1.049.4%estimated ± 3.2 pp, medium confidence
SEC-Bench Pro42.5%estimated ± 9.3 pp, medium confidence
ExploitBench Internal Port0.0%estimated ± 1.4 pp, low confidence