Financial Question Answering on FinMMR Hard
50.8AccuracyClaude Sonnet
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude SonnetGroup=General LLMs2026.06 | 50.8 | |
| LlamaGroup=General LLMs2026.06 | 48.7 | |
| Qwen3-VL-8B + FINACUMENGroup=FINACUMEN2026.06 | 46.1 | |
| GPTGroup=General LLMs2026.06 | 45.4 | |
| Gemini 2.0 FlashGroup=General LLMs2026.06 | 44.4 | |
| Qwen2.5-VL-72BGroup=General LLMs2026.06 | 43.3 | |
| Qwen3-VL-8B-InstructGroup=Base2026.06 | 31.4 | |
| FinLLaVAGroup=Financial LLMs2026.06 | 3.9 |