Reasoning Models Ace the CFA Exams

TL;DR

This study evaluates state-of-the-art reasoning models on mock CFA exams, demonstrating that most models can pass all levels with high accuracy, surpassing previous LLM performance on financial certification tests.

Contribution

The paper provides the first comprehensive assessment of advanced reasoning models on CFA exams, showing their ability to pass all levels with high scores, unlike prior LLMs.

Findings

Most models pass all CFA levels with high accuracy.

Gemini 3.0 Pro achieves 97.6% on Level I.

GPT-5 scores 94.3% on Level II.

Abstract

Previous research has reported that large language models (LLMs) demonstrate poor performance on the Chartered Financial Analyst (CFA) exams. However, recent reasoning models have achieved strong results on graduate-level academic and professional examinations across various disciplines. In this paper, we evaluate state-of-the-art reasoning models on a set of mock CFA exams consisting of 980 questions across three Level I exams, two Level II exams, and three Level III exams. Using the same pass/fail criteria from prior studies, we find that most models clear all three levels. The models that pass, ordered by overall performance, are Gemini 3.0 Pro, Gemini 2.5 Pro, GPT-5, Grok 4, Claude Opus 4.1, and DeepSeek-V3.1. Specifically, Gemini 3.0 Pro achieves a record score of 97.6% on Level I. Performance is also strong on Level II, led by GPT-5 at 94.3%. On Level III, Gemini 2.5 Pro attains the highest score with 86.4% on multiple-choice questions while Gemini 3.0 Pro achieves 92.0% on constructed-response questions.