Google / MODEL EVIDENCE
49 benchmarks · 71 observations · 53 sources · released 2026-07-21
49 benchmarks · large dot: Gemini 3.6 Flash; small dots: closest models where reported
Gemini 3.6 Flash · GPQA diamond · Best score (across scorers) · % ↗
86.36% · Gemini 3.6 Flash · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
85.86% · Gemini 3.6 Flash · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
93.43 % · Gemini 3.6 Flash · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗
92.83% · Gemini 3.6 Flash (High) · GPQA Diamond · AA implementation · Artificial Analysis ↗
Gemini 3.6 Flash · MMMU Pro · Accuracy ↗
83.24% · Gemini 3.6 Flash (High) · MMMU-Pro · AA implementation · Artificial Analysis ↗
Gemini 3.6 Flash · FrontierMath-Tiers-1-3-v2-Private · Best score (across scorers) · % ↗
Gemini 3.6 Flash · FrontierMath-Tier-4-v2-Private · Best score (across scorers) · % ↗
Gemini 3.6 Flash · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
82.22% · Gemini 3.6 Flash · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % · Epoch AI · OTIS Mock AIME 2024-2025 ↗
80% · Gemini 3.6 Flash · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % · Epoch AI · OTIS Mock AIME 2024-2025 ↗
Gemini 3.6 Flash · SimpleQA Verified · Best score (across scorers) · % ↗
Gemini 3.6 Flash · Chess Puzzles · Best score (across scorers) · % ↗
40% · Gemini 3.6 Flash · Chess Puzzles · Best score (across scorers) · % · Epoch AI · Chess Puzzles ↗
29% · Gemini 3.6 Flash · Chess Puzzles · Best score (across scorers) · % · Epoch AI · Chess Puzzles ↗
Gemini 3.6 Flash · Mystery Game Puzzles · Best score (across scorers) · % ↗
25% · Gemini 3.6 Flash · Mystery Game Puzzles · Best score (across scorers) · % · Epoch AI · Mystery Game Puzzles ↗
23% · Gemini 3.6 Flash · Mystery Game Puzzles · Best score (across scorers) · % · Epoch AI · Mystery Game Puzzles ↗
Gemini 3.6 Flash · ARC-AGI · Score · % ↗
76.5% · Gemini 3.6 Flash · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
34.5% · Gemini 3.6 Flash · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
83.17% · Gemini 3.6 Flash (medium) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
Gemini 3.6 Flash · WeirdML · Accuracy · % ↗
Gemini 3.6 Flash · APEX-Agents · Pass@1 score · % ↗
Gemini 3.6 Flash · ARC-AGI-2 · Score · % ↗
30.42% · Gemini 3.6 Flash · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
2.64% · Gemini 3.6 Flash · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
50.42% · Gemini 3.6 Flash (medium) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
Gemini 3.6 Flash · CritPt · Accuracy · % ↗
Gemini 3.6 Flash · DeepSWE · Pass@1 · % ↗
Gemini 3.6 Flash · ProofBench · Accuracy · % ↗
Gemini 3.6 Flash · FrontierCode · Main score · % ↗
Gemini 3.6 Flash · LMCA · Score · % ↗
Gemini 3.6 Flash · DTBench · Accuracy · % ↗
Gemini 3.6 Flash · Furniture Assembly · Best score (across scorers) · % ↗
Gemini 3.6 Flash · GDP.pdf · GDP.pdf score · % ↗
Gemini 3.6 Flash · SciCode · Score · % ↗
Gemini 3.6 Flash · WebDev Arena · Elo ↗
1527.83 Elo · Gemini 3.6 Flash · WebDev Arena · Elo · Epoch AI · WebDev Arena ↗
Gemini 3.6 Flash · Blueprint-Bench 2 · Score · % ↗
Gemini 3.6 Flash · General ECI · 26 October 2026 fit ↗
Gemini 3.6 Flash · Vals Multimodal Index · Accuracy ↗
Gemini 3.6 Flash · Legal Research Bench · All-Pass ↗
Gemini 3.6 Flash · LegalBench · Accuracy ↗
Gemini 3.6 Flash · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Gemini 3.6 Flash · Tax Agent Bench · Accuracy ↗
Gemini 3.6 Flash · Finance Agent v2 · Accuracy ↗
Gemini 3.6 Flash · Excel Modeling Benchmark v1 · Accuracy ↗
Gemini 3.6 Flash · CorpFin · Accuracy ↗
Gemini 3.6 Flash · MortgageTax · Accuracy ↗
74.86 % · Gemini 3.6 Flash · TaxEval · Accuracy · Vals AI · TaxEval ↗
Gemini 3.6 Flash · MedCode · Accuracy ↗
Gemini 3.6 Flash · MedScribe · Accuracy ↗
Gemini 3.6 Flash · Urgent safety · share of applicable ratings marked fail ↗
68% · Gemini 3.6 Flash · Health advice · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗
70% · Gemini 3.6 Flash · Relationship boundaries · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗
Gemini 3.6 Flash (Mini-SWE-agent) · Terminal-Bench-Science · Score ↗
Gemini 3.6 Flash · BioMysteryBench · Accuracy ↗
Gemini 3.6 Flash · MMLU Pro · Accuracy ↗
Gemini 3.6 Flash · SAGE · Accuracy ↗
Gemini 3.6 Flash (OpenHands) · Vibe Code Bench · original task set ↗
Gemini 3.6 Flash · ProgramBench · Accuracy ↗
Gemini 3.6 Flash · IOI · Accuracy ↗
Gemini 3.6 Flash · Code Migration v1 · Accuracy ↗
Gemini 3.6 Flash (Mini-SWE-agent) · SWE-bench Verified · Score ↗
Gemini 3.6 Flash · LiveCodeBench · Accuracy ↗
Gemini 3.6 Flash · CyberBench · Accuracy ↗
Gemini 3.6 Flash · Public Benefits Bench · Accuracy ↗
Gemini-3.6-Flash (high) · Humanity's Sixth Sense · Score ↗
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
21 of 30 shared benchmark comparisons
21 of 30 shared benchmark comparisons
20 of 30 shared benchmark comparisons
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
Gemini 3.6 Flash · Gemini 3.6 Flash (High) · Gemini 3.6 Flash (Mini-SWE-agent) · Gemini 3.6 Flash (OpenHands) · Gemini 3.6 Flash (medium) · Gemini-3.6-Flash (high)