FINESSE-Bench - CFA-like Level 2: leaderboard

FINESSE-Bench CFA-like Level 2 tests financial knowledge with 293 multiple-choice item sets on applied valuation, financial statement analysis, fixed income and derivatives from FINESSE-Bench (2026). Answers are marked correct or incorrect by a GPT-5.2 judge against the reference; the score is accuracy in percent, higher is better.

Metric: Accuracy (%) on the 293 CFA-like Level 2 multiple-choice item sets on applied valuation, financial statement analysis, fixed income and derivatives of FINESSE-Bench; a GPT-5.2 judge marks each answer correct or incorrect against the reference; zero-shot, one fixed prompt per task type, temperature 0 where possible, reasoning configurations with medium effort where the model offers them; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 31 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Kimi K2.591.81#266
2Claude Sonnet 4.6Medium91.81#173
3GLM-4.791.47#380
4Qwen 3.5 Plus (2026-02-15)91.47#194
5GLM-589.76#269
6Qwen 3.5 27B89.08#375
7Qwen 3.5 122B A10B88.74#346
8GPT-5.2Medium88.74#185
9Qwen 3.5 Flash (02-23)88.74#557
10Qwen 3.5 35B A3B88.4#492
11Qwen 3 235B A22B 2507 (Thinking)86.35#466
12DeepSeek R1 052885.32#428
13Claude 3.7 SonnetThinking83.96#470
14Qwen 3.5 397B A17B82.25#264
15MiniMax-M2.581.57#600

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=finesse-bench-cfa-like-level-2 · How It Works · Data refreshed daily, snapshot 2026-11-21.