CFLUE: leaderboard
Chinese Financial Language Understanding Evaluation benchmark for assessing LLMs across financial knowledge, reasoning, and domain-language tasks.
Metric: Accuracy. Source: github.com. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen-72B | 72.8 | #779 |
| 2 | GPT-4 | 60.87 | #438 |
| 3 | GPT-4 Turbo | 60.61 | #404 |
| 4 | Qwen-14B | 53.82 | #1171 |
| 5 | Qwen-7B | 43.63 | #1402 |
| 6 | ChatGPT | 43.35 | #900 |
| 7 | chatglm3-6B | 40.78 | #1439 |
| 8 | Qwen-1.8B | 38.68 | #1546 |
| 9 | Llama 2 7B | 18.79 | #1520 |
| 10 | Llama 2 70B | 17.66 | #1201 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=cflue · How It Works · Data refreshed daily, snapshot 2026-10-20.