Financial Question Answering on BizBench SEC-NUM
79.3AccuracyGPT
Evaluation Results
| Method | Links | |
|---|---|---|
| GPTGroup=General LLMs2026.06 | 79.3 | |
| LlamaGroup=General LLMs2026.06 | 74.7 | |
| Qwen2.5-VL-72BGroup=General LLMs2026.06 | 69.85 | |
| Qwen3-VL-8B + FINACUMENGroup=FINACUMEN2026.06 | 68.65 | |
| Llama-Open-Finance-8BGroup=Financial LLMs2026.06 | 63.64 | |
| Claude SonnetGroup=General LLMs2026.06 | 63.39 | |
| Fin-R1Group=Financial LLMs2026.06 | 59.32 | |
| FinLLaVAGroup=Financial LLMs2026.06 | 53.46 | |
| Qwen3-VL-8B-InstructGroup=Base2026.06 | 27.67 |