Financial Reasoning on ConvFinQA
85.7Accuracy
GPT-5 mini-high
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 85.7 | 77.1 | |
2026.03 | 82.5 | 74.7 | |
2026.03 | 80.6 | 72.5 | |
2026.03 | 80.4 | 74.6 | |
2026.03 | 79.5 | 67.7 | |
2026.03 | 78.8 | 71.5 | |
2026.03 | 78.3 | 72.1 | |
2026.03 | 76.2 | 61.4 | |
2026.03 | 74.6 | 70.3 | |
2026.03 | 72.1 | 68 | |
2026.03 | 69.6 | 47.8 | |
2026.03 | 68.2 | 56.9 | |
2026.01 | 55.85 | - | |
2026.01 | 51.01 | - | |
2026.01 | 50.35 | - | |
2026.01 | 36.54 | - | |
2026.01 | 34.02 | - | |
2026.01 | 18.87 | - | |
2026.01 | 18.44 | - | |
2026.01 | 17.75 | - | |
2026.01 | 17.43 | - | |
2026.03 | 10.9 | 36.7 | |
2026.03 | 9.3 | 36.2 |