# Gemini 3.7 Flash - BenchAudit True Index Profile

**Integrity Profile: High Contamination Vulnerability**

> The Verdict: Strong on Public Leaderboards, Vulnerable to Fresh Evals. Gemini 3.7 Flash (evaluated at high effort) shows a wide divergence between public benchmark claims (42.9p) and private evaluation performance (39.4p), with a +19.2p Low-Tier Boost across 3 matched domains, especially in General Knowledge & QA (+16.9p boost). Treat legacy leaderboard scores cautiously and anchor decisions on its Top-Tier S & A results.

- **Canonical URL:** https://benchaudit.org/models/gemini-3-7-flash__high/
- **True Index Rank:** #31 of 44
- **Creator:** Google
- **Weights:** Closed / API
- **Clean Evals Avg (Tiers S & A):** 39.4p (8 evals)
- **Legacy Benchmarks Avg (Tiers C & D):** 42.9p (4 evals)
- **Low-Tier Boost (Matched Delta):** +19.2p
- **Updated:** 2026-11-22

## True Index: Domain Rankings

| Capability Domain | Sub-Rank | Clean Evals (S/A) Avg | Top-Tier Evals |
| --- | ---: | ---: | ---: |
| Agentic & Enterprise | #22 | 49.6p | 3 |
| Coding & SWE | #33 | 28.6p | 1 |
| Math & Scientific Reasoning | #37 | 26.5p | 3 |

## Head-to-Head: Clean Evals vs. Legacy Benchmarks

| Capability Domain | Clean Evals (Tiers S & A) | Legacy Benchmarks (Tiers C & D) | Public Lift (Boost) |
| --- | ---: | ---: | ---: |
| Agentic & Enterprise | 3 (49.6p) | 2 (36.4p) | -13.2p |
| Coding & SWE | 1 (28.6p) | 1 (22.4p) | -6.2p |
| General Knowledge & QA | 1 (53.0p) | 1 (69.9p) | +16.9p |

## Audited Benchmark Performance of Gemini 3.7 Flash

| Benchmark | Benchmark Integrity Status | Domain | Rank | Percentile |
| --- | --- | --- | ---: | ---: |
| [AA-AnalystAgent](https://benchaudit.org/benchmarks/aa-analystagent/) | Private / Fresh | Agentic & Enterprise | #2/24 | 93.7p |
| [MMMU-Pro](https://benchaudit.org/benchmarks/mmmu-pro/) | Public / Aging | General Knowledge & QA | #6/22 | 69.9p |
| [GDP.pdf](https://benchaudit.org/benchmarks/gdp-pdf/) | Public / Aging | Agentic & Enterprise | #20/50 | 61.2p |
| [ARC-AGI-2](https://benchaudit.org/benchmarks/arc-agi-2/) | Private / Fresh | General Knowledge & QA | #13/22 | 53.0p |
| [AutomationBench-AA](https://benchaudit.org/benchmarks/automationbench-aa/) | Private / Fresh | Agentic & Enterprise | #26/50 | 49.0p |
| [FrontierMath](https://benchaudit.org/benchmarks/frontiermath/) | Private / Fresh | Math & Scientific Reasoning | #20/27 | 42.5p |
| [CyberBench v1.1](https://benchaudit.org/benchmarks/cyberbench/) | Private / Fresh | Coding & SWE | #21/25 | 28.6p |
| [Terminal-Bench 4.0](https://benchaudit.org/benchmarks/terminal-bench-4-0/) | Public / Aging | Coding & SWE | #39/50 | 22.4p |
| [CritPt v2](https://benchaudit.org/benchmarks/critpt-v2/) | Private / Fresh | Math & Scientific Reasoning | #30/40 | 21.2p |
| [MLCR-AA](https://benchaudit.org/benchmarks/mlcr-aa/) | Private / Fresh | Math & Scientific Reasoning | #41/46 | 15.8p |
| [GDPval-AA v2.1](https://benchaudit.org/benchmarks/gdpval-aa-v2-1/) | Public / Aging | Agentic & Enterprise | #37/43 | 11.5p |
| [AA-Briefcase v1.1](https://benchaudit.org/benchmarks/aa-briefcase-v1-1/) | Private / Fresh | Agentic & Enterprise | #47/50 | 6.1p |
