# Gemini 2.5 Flash vs o4-mini

Every finance benchmark where both have a score (Google and OpenAI). Data refreshed 2 Nov 2026. Scores from different benchmarks use different scales and are not added up.

| Benchmark | Metric | Gemini 2.5 Flash | Gemini 2.5 Flash source | o4-mini | o4-mini source | Ahead |
| --- | --- | --- | --- | --- | --- | --- |
| Advanced Financial Reasoning at Scale (CFA Level III evaluation) | overall % (mean of MCQ and essay) (higher is better) | 77.3 | Paper Table 1, 22 Sept 2025, https://arxiv.org/html/2507.02954v2 | 79.1 | Paper Table 1, 22 Sept 2025, https://arxiv.org/html/2507.02954v2 | o4-mini |
| FinTrust | safety harmfulness score (1-10, GPT-4.1 mini judge) (lower is better) | 6.2 | Paper Table 7 (App. A.2), 17 Oct 2025, https://arxiv.org/html/2510.15232v1 | 1 | Paper Table 7 (App. A.2), 17 Oct 2025, https://arxiv.org/html/2510.15232v1 | o4-mini |

Source: https://benchmarks.finance/compare/gemini-2-5-flash-vs-o4-mini/

Created and maintained by Moneda.