BizFinBench — leaderboard
Financial benchmark evaluating LLMs across 9 business finance tasks: anomalous event attribution, financial computation, time reasoning, tool usage, QA, data description, emotion recognition, stock prediction, and named entity recognition.
Metric: Average Score. Source: github.com. Status: saturation imminent. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek R1 | 73.05 |
| 2 | GPT-4o | 71.8 |
| 3 | DeepSeek V3 | 71.57 |
| 4 | Gemini 2.0 Flash | 69.75 |
| 5 | Qwen 3 32B | 68.26 |
| 6 | Qwen 2.5 72B Instruct | 67.7 |
| 7 | Qwen 3 14B | 67.05 |
| 8 | DeepSeek R1 Distill Qwen 32B | 66.29 |
| 9 | QwQ-32B | 65.77 |
| 10 | Claude 3.5 Sonnet | 65.59 |
| 11 | Llama 4 Scout | 61.17 |
| 12 | Qwen 3 4B | 59.94 |
| 13 | DeepSeek R1 Distill Qwen 14B | 59.49 |
| 14 | Qwen 2.5 7B Instruct | 56.35 |
| 15 | Llama 3.1 70B Instruct | 55.09 |
Interactive version: theaggregate.ai/benchmark?slug=bizfinbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.