FinEval - Financial Academic Knowledge (Zero-shot CoT): leaderboard
FinEval's financial academic knowledge section: Chinese multiple-choice questions on finance, economics, accounting and professional certificate exams. Scored as accuracy (%), averaged over subjects, under zero-shot chain-of-thought prompting; higher is better.
Metric: Accuracy (%, zero-shot chain-of-thought). Source: github.com. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 76.5 | #1093 |
| 2 | Claude 3.5 Sonnet | 74.4 | #747 |
| 3 | GPT-4o | 72.1 | #741 |
| 4 | Qwen 2.5 7B Instruct | 69.7 | #2583 |
| 5 | Gemini 1.5 Pro | 68.9 | #1053 |
| 6 | internlm2.5-20B-chat | 67.3 | #2563 |
| 7 | GLM-4 9B Chat | 66.6 | #2782 |
| 8 | Yi 1.5 34B Chat | 65 | #2513 |
| 9 | Yi-1.5-9B Chat | 62.7 | #4060 |
| 10 | GPT-4o Mini | 61.9 | #1627 |
| 11 | Gemini 1.5 Flash | 61.2 | #1671 |
| 12 | internlm2-chat-20B | 60.3 | #2896 |
| 13 | Baichuan2-13B-Chat | 47.7 | #6615 |
| 14 | chatglm3-6B | 44.3 | #8269 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=fineval-financial-academic-knowledge-zero-shot-cot · How It Works · Data refreshed daily, snapshot 2026-11-12.