Model

Gemini 3.7 Flash

Google

deepmind.google

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
ArcticAbstain7.3abstention_rateN/AIndependentmediumarxiv.org
CritPt14.3accuracyN/AIndependentlowepoch.ai
GPQA Diamond94.8accuracyN/AIndependenthighepoch.ai
MOF-VERIFY62.8macro-f1N/AIndependentmediumarxiv.org
ScholarCatalyst0.4r@20tool-calling agentIndependentmediumarxiv.org
SimpleQA Verified69.2accuracyN/AIndependenthighepoch.ai

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
DeepSWE65.5pass@1mini-swe-agentIndependentmediumdeepswe.datacurve.ai
MetaKernelBench59.5paired_liftcute dslIndependentmediumarxiv.org
SciCode59.8scoreN/AIndependentmediumartificialanalysis.ai
SWE-CC80.2resolve_ratemini-swe-agentIndependentmediumarxiv.org
WorldSolver29.3scoreN/AIndependentmediumarxiv.org

Agentic / Tool Use

Agentic / Tool Use
BenchmarkScoreMetricHarnessSourceConfidenceLink
cua-speedrun89.6success_rateN/AIndependentmediumarxiv.org
DUDA-Bench0.2workflow_scorereactIndependentmediumarxiv.org
EngiWorld25.3engiscoreN/AIndependentmediumarxiv.org
HyperBrowseComp31.7accuracyN/AIndependentmediumarxiv.org
HyperBrowseComp17.7accuracyowlIndependentmediumarxiv.org
KlinikeBench10.8strict_pass@1N/AIndependentmediumarxiv.org

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
DSV-Mem39.6accuracyN/AIndependentmediumarxiv.org
Ego-SpaCR41.4accuracynative videoIndependentmediumarxiv.org
GDP.pdf23.8gdp.pdf_scoreN/AIndependentmediumsurgehq.ai
HSS39.8pass@1N/AIndependentmediumarxiv.org
NextAct0.9h_10N/AIndependentmediumarxiv.org
PlotGround-1k65.7accuracyN/AIndependentmediumarxiv.org

Safety & Alignment

Safety & Alignment
BenchmarkScoreMetricHarnessSourceConfidenceLink
AgentDoxx97.2accuracyN/AIndependentmediumarxiv.org

Abstract Reasoning

Abstract Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
ARC-AGI-284.6scoreN/AIndependentlowepoch.ai

Reproduced by the community

Loading community runs…