CFMME: leaderboard
The emergence of Large Vision-Language Models (LVLMs) has substantially expanded model capabilities beyond text-only understanding, enabling unified inference across both visual and textual modalities and supporting a broader range of real-world applications. To comprehensively evaluate the perception, understanding, reasoning, and cognition capabilities of LVLMs throughout the entire financial bu
Metric: Average (self-reported). Source: benchmarklist.com. Saturation forecast: Around December 2026. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B (Thinking) | 66.11 | #238 (Qwen 3 VL 235B A22B) |
| 2 | Qwen 3 VL 235B A22B Instruct | 63.31 | #273 |
| 3 | Qwen 3 VL 8B (Thinking) | 53.85 | |
| 4 | Qwen 3 VL 8B Instruct | 52.83 | #420 |
| 5 | GLM-4.1V-9B (Thinking) | 51.51 | #472 (GLM-4.1V-9B) |
| 6 | GPT-4o | 46.96 | #343 |
| 7 | ERNIE 4.5 VL 28B A3B | 45.68 | #695 |
| 8 | Llama 3.2 11B Instruct | 20.2 | #1126 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=cfmme · How It Works · Data refreshed daily, snapshot 2026-10-20.