FinEval - Financial Agent: leaderboard
Metric: Similarity score (%, zero-shot). Source: github.com. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet | 79.3 |
| 2 | GPT-4o | 73.9 |
| 3 | GPT-4o Mini | 72.9 |
| 4 | Gemini 1.5 Pro | 72.8 |
| 5 | Gemini 1.5 Flash | 70.9 |
| 6 | Qwen 2.5 72B Instruct | 68.4 |
| 7 | Qwen 2.5 7B Instruct | 66.7 |
| 8 | Yi 1.5 34B Chat | 66 |
| 9 | internlm2.5-20B-chat | 63.1 |
| 10 | Yi-1.5-9B Chat | 61.1 |
| 11 | internlm2-chat-20B | 60.9 |
| 12 | GLM-4 9B Chat | 60.2 |
| 13 | Baichuan2-13B-Chat | 55.7 |
| 14 | chatglm3-6B | 49.6 |
Interactive version: theaggregate.ai/benchmark?slug=fineval-financial-agent · How It Works · Data refreshed daily, snapshot 2026-09-26.