IPO Finance Agent Performance Evaluation on SpaceX Question Set (n=70 public)
79.4Mean ScoreQwen 3.7 Max
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen 3.7 MaxCost / query=$0.302026.06 | 79.4 | 36.4 | 100 | |
| GLM-5.12026.06 | 77.6 | 48.5 | 100 | |
| MiMo-2.5 ProCost / query=$0.052026.06 | 76.8 | 28.3 | 100 | |
| Kimi K2.6Cost / query=$0.212026.06 | 67.2 | 0 | 100 | |
| Nemotron 3 UltraCost / query=$0.142026.06 | 43.9 | 0 | 100 |