benchgap
DeepSeek · model

DeepSeek V3.2 benchmark scores

As of 2026-11-22, DeepSeek V3.2 (DeepSeek) has measured scores on 16 benchmarks and estimated scores on 77 more.

BenchmarkScoreSource
AA-GPQA Diamond75.1%measured
AA-HLE11.2%measured
AA-SciCode45.4%estimated ± 8.6 pp, low confidence
Artificial Analysis Intelligence Index16.0%measured
CritPt0.9%measured
GDPval-AA15.9%estimated ± 10.5 pp, low confidence
GPQA Diamond74.5%estimated ± 2.4 pp, high confidence
HLE18.8%estimated ± 8.5 pp, medium confidence
IFBench55.6%estimated ± 6.6 pp, medium confidence
AA-LCR45.7%measured
AA-Omniscience Accuracy24.0%measured
BrowseComp50.9%estimated ± 8.0 pp, low confidence
HLE w/ tools30.5%estimated ± 5.5 pp, low confidence
IFEval90.9%estimated ± 2.8 pp, high confidence
LongBench v251.8%estimated ± 5.5 pp, medium confidence
VITA-Bench18.5%measured
τ²-bench results78.9%measured
AA Coding Index40.1%estimated ± 11.1 pp, low confidence
APEX-Agents-AA9.6%estimated ± 9.3 pp, medium confidence
BioMysteryBench (human-difficult)15.7%estimated ± 6.8 pp, low confidence
SWE-bench Verified78.0%estimated ± 4.8 pp, medium confidence
BFCL v460.8%estimated ± 8.5 pp, low confidence
CyberGym42.2%estimated ± 11.0 pp, low confidence
JobBench0.0%estimated ± 9.3 pp, low confidence
WideResearch63.1%estimated ± 3.3 pp, medium confidence
AA Agentic Index12.5%estimated ± 9.4 pp, medium confidence
MMLU-Pro84.9%estimated ± 5.0 pp, medium confidence
AA-IFBench49.0%measured
FailureSensorIQ62.0%estimated ± 1.7 pp, medium confidence
GPQA77.0%estimated ± 6.1 pp, medium confidence
MMLU90.4%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond74.2%estimated ± 3.2 pp, high confidence
Vals MMLU-Pro82.0%estimated ± 2.9 pp, high confidence
MMMLU86.3%estimated ± 0.7 pp, medium confidence
Chess Puzzles19.3%estimated ± 12.3 pp, low confidence
AA MMLU-Pro84.7%estimated ± 0.8 pp, high confidence
Gert Labs21.1%measured
ApprenticeBench3.1%estimated ± 10.4 pp, low confidence
ARC-AGI-175.4%estimated ± 7.8 pp, low confidence
ARC-AGI-232.3%estimated ± 14.4 pp, low confidence
MazeBench0.0%estimated ± 2.7 pp, medium confidence
Mystery Game Puzzles14.1%estimated ± 2.2 pp, low confidence
NYT Connections8.2%measured
OSWorld-Verified38.7%estimated ± 3.8 pp, low confidence
Vals SWE-bench63.5%estimated ± 7.0 pp, low confidence
CWE-Bench-AA32.3%estimated ± 8.1 pp, low confidence
AA Cyber Index4.3%estimated ± 12.4 pp, low confidence
DeepsecBench-AA6.2%estimated ± 8.2 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)50.7%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)51.2%estimated ± 1.1 pp, low confidence
AA ITBench35.7%estimated ± 4.6 pp, low confidence
CWE-bench v133.7%estimated ± 5.7 pp, low confidence
HLE w/o tools28.7%estimated ± 4.9 pp, high confidence
OSWorld 2.11.6%estimated ± 3.6 pp, medium confidence
StealthMark41.3%estimated ± 10.7 pp, low confidence
Toolathlon-Verified55.8%estimated ± 3.0 pp, high confidence
AA AutomationBench38.4%estimated ± 9.9 pp, low confidence
GDP.pdf10.0%estimated ± 4.2 pp, medium confidence
ARC-AGI-31.3%estimated ± 9.8 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
CWE-bench26.5%estimated ± 6.2 pp, low confidence
MKA32.8%estimated ± 4.6 pp, medium confidence
SkillsBench official leaderboard (BenchFlow)19.2%estimated ± 6.1 pp, medium confidence
BinaryAudit12.3%measured
OTelBench11.6%measured
Vibe Code Bench27.7%estimated ± 6.5 pp, low confidence
Hermes Index15.0%estimated ± 5.7 pp, low confidence
ExploitGym0.0%estimated ± 8.3 pp, low confidence
C-Eval90.4%estimated ± 1.0 pp, medium confidence
Claw-Eval40.2%measured
DeepPlanning26.2%estimated ± 9.1 pp, low confidence
MCP Atlas40.9%estimated ± 13.3 pp, low confidence
MMLU-Redux94.0%estimated ± 0.8 pp, medium confidence
QwenClawBench52.0%estimated ± 2.0 pp, low confidence
SuperGPQA65.5%estimated ± 9.3 pp, medium confidence
Toolathlon37.8%estimated ± 8.0 pp, medium confidence
HealthBench Hard8.9%estimated ± 8.6 pp, low confidence
MMLU-Pro (Arcee)80.5%estimated ± 4.0 pp, medium confidence
React Native Evals71.5%measured
ResearchClawBench11.3%estimated ± 1.3 pp, medium confidence
SWE-Rebench60.9%measured
Finance Agent v255.1%estimated ± 2.4 pp, low confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
HealthBench (length-adjusted)54.1%estimated ± 3.7 pp, medium confidence
HealthBench Professional56.7%estimated ± 4.2 pp, low confidence
HLE-Verified18.9%estimated ± 7.1 pp, low confidence
STEMBench2.4%estimated ± 0.2 pp, low confidence
SimpleQA23.6%estimated ± 4.3 pp, medium confidence
Agents' Last Exam29.6%estimated ± 9.7 pp, low confidence
MRCRv222.8%estimated ± 11.3 pp, low confidence
MMLongBench-Doc59.0%estimated ± 2.6 pp, low confidence
BioBasicsBench34.1%estimated ± 3.8 pp, low confidence
LifeVal0.9%estimated ± 5.0 pp, low confidence