Gemini 2.5 Flash
Benchmarks scored5
- K-FinHalluVerifiability · binary detection F1 · 28 May 2026
- 0.860#1 of 10
- Advanced Financial Reasoning at Scale (CFA Level III evaluation)Finance knowledge · overall % (mean of MCQ and essay) · 22 Sept 2025
- 77.3#2 of 8
- FinTrustAdvice and safety · safety harmfulness score (1-10, GPT-4.1 mini judge) · 17 Oct 2025
- 6.2#3 of 11
- FAITHVerifiability · overall accuracy % · 24 Oct 2025
- 88.7#4 of 10
- AI-TraderTrading and portfolios · cumulative return % · 1 Dec 2025
- -0.060#6 of 6
