| Models,FinQA,DM-Simplong,XBRL-Math,Average,Model types | |
| GPT-4o,72.49,60.0,72.22,68.24,instruction-tuned | |
| GPT-o1,49.07,56.0,74.44,59.84,instruction-tuned | |
| GPT-o3-mini,60.87,59.0,76.67,65.51,instruction-tuned | |
| DeepSeek-V3,73.2,53.0,76.67,67.62,instruction-tuned | |
| DeepSeek-R1,65.13,53.0,86.67,68.93,instruction-tuned | |
| Qwen2.5-72B-Instruct,73.38,59.0,67.78,66.72,instruction-tuned | |
| Qwen2.5-72B-Instruct-Math,69.74,42.0,83.33,65.69,instruction-tuned | |
| DeepSeek-R1-Distill-Llama-70B,66.73,53.0,86.67,68.8,instruction-tuned | |
| Llama3-70B-Instruct,58.92,41.0,56.67,52.2,instruction-tuned | |
| Llama3.1-70B-Instruct,63.18,48.0,63.33,58.17,instruction-tuned | |
| Llama3.3-70B-Instruct,68.15,54.0,70.0,64.05,instruction-tuned | |
| DeepSeek-R1-Distill-Qwen-32B,65.48,55.0,84.44,68.97,instruction-tuned | |
| DeepSeek-R1-Distill-Qwen-14B,63.27,44.0,84.44,63.9,instruction-tuned | |
| DeepSeek-R1-Distill-Llama-8B,45.96,33.0,81.11,53.36,instruction-tuned | |
| Llama3-8B-Instruct,41.97,29.0,48.89,39.95,instruction-tuned | |
| Llama3.1-8B-Instruct,54.13,34.0,62.22,50.12,instruction-tuned | |
| Fino1-8B,60.87,40.0,82.22,61.03,instruction-tuned | |