evals.report
BenchmarksLabsCompareRun guidesIn the wild

Gemini 3.7 Flash

Google DeepMind · Gemini. Released Aug 13, 2026.

Gemini 3.7 Flash is a model from Google DeepMind in the Gemini family, released Aug 13, 2026. evals.report tracks 6 reported Gemini 3.7 Flash benchmark scores across ARC-AGI-1, ARC-AGI-2, FrontierCode, Terminal-Bench 2.1, DeepSWE, CharXiv — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.

6 results

Benchmark results 6

Compare this model
BenchmarkCategoryScoreMetricStatusDate
ARC-AGI-1Reasoning95.5%accuracyOfficialAug 13, 2026Details
ARC-AGI-2Reasoning84.58%accuracyOfficialAug 13, 2026Details
FrontierCodeCoding43.6%weighted score (Main)VerifiedAug 13, 2026Details
Terminal-Bench 2.1Agents85.8%task successVerifiedAug 13, 2026Details
DeepSWECoding65.3%% resolvedVerifiedAug 13, 2026Details
CharXivMultimodal84.5%accuracyVerifiedAug 13, 2026Details