FinEval - Financial Academic Knowledge (Zero-shot): leaderboard
Metric: Accuracy (%, zero-shot). Source: github.com. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet | 73.9 |
| 2 | GPT-4o | 71.5 |
| 3 | Qwen 2.5 72B Instruct | 69.7 |
| 4 | Gemini 1.5 Pro | 68.3 |
| 5 | Qwen 2.5 7B Instruct | 62.7 |
| 6 | GPT-4o Mini | 62.4 |
| 7 | Gemini 1.5 Flash | 62.1 |
| 8 | Yi 1.5 34B Chat | 59.5 |
| 9 | Yi-1.5-9B Chat | 55 |
| 10 | GLM-4 9B Chat | 54.7 |
| 11 | internlm2-chat-20B | 54.7 |
| 12 | internlm2.5-20B-chat | 54.7 |
| 13 | Baichuan2-13B-Chat | 41.1 |
| 14 | chatglm3-6B | 38.9 |
Interactive version: theaggregate.ai/benchmark?slug=fineval-financial-academic-knowledge-zero-shot · How It Works · Data refreshed daily, snapshot 2026-09-26.