CFMME - Application QA: leaderboard
Metric: Accuracy (%) on question answering over real financial document images (application assessment) on CFMME (6,052 Chinese financial multimodal instances over eight financial image types), zero-shot, mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 235B A22B (Thinking) | 54.94 |
| 2 | Qwen 3 VL 235B A22B Instruct | 52.57 |
| 3 | GLM-4.5V (Non-reasoning) | 50.13 |
| 4 | Qwen 3 VL 8B (Thinking) | 44.22 |
| 5 | Qwen 3 VL 8B Instruct | 43.8 |
| 6 | ERNIE 4.5 VL 28B A3B | 42.07 |
| 7 | GLM-4.1V-9B (Thinking) | 41.81 |
| 8 | GPT-4o | 37.05 |
| 9 | Llama 3.2 11B Instruct | 18.61 |
Interactive version: theaggregate.ai/benchmark?slug=cfmme-application-qa · How It Works · Data refreshed daily, snapshot 2026-10-07.