Multi-Constraint Scheduling on NaturalPlan
48.6Trip Success RateMagicAgent-32B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MagicAgent-32BModel Category=MagicAgent Series2026.02 | 48.6 | 57.7 | 61.5 | |
| MagicAgent-30B-A3BModel Category=MagicAgent Series2026.02 | 42.8 | 60.7 | 58.3 | |
| GPT-5.2Model Category=Ultra-Scale2026.02 | 39.4 | 53.7 | 64.5 | |
| Qwen3-32B-nothinkModel Category=Large-Scale2026.02 | 37.6 | 21.5 | 31.9 | |
| GLM-4.7Model Category=Ultra-Scale2026.02 | 31.3 | 42.7 | 46.4 | |
| Kimi-K2-InstructModel Category=Ultra-Scale2026.02 | 30.9 | 51.4 | 25.7 | |
| Llama3.3-70B-InstructModel Category=Large-Scale2026.02 | 30.8 | 33.8 | 44 | |
| Qwen3-235B-A22B-InstructModel Category=Ultra-Scale2026.02 | 21.9 | 31.7 | 53.5 | |
| Qwen3-30B-A3B-InstructModel Category=Large-Scale2026.02 | 19.4 | 20.3 | 38.4 | |
| Olmo-3.1-32B-InstructModel Category=Large-Scale2026.02 | 14.1 | 20.2 | 26.2 | |
| ERNIE-4.5-21B-A3B-PTModel Category=Large-Scale2026.02 | 5.9 | 11 | 16.3 | |
| GLM-4.7-FlashModel Category=Large-Scale2026.02 | 5.9 | 16.9 | 21.3 | |
| DeepSeek-V3.1-nothinkModel Category=Ultra-Scale2026.02 | 1.4 | 37.2 | 64.4 | |
| Qwen3-MAXModel Category=Ultra-Scale2026.02 | 1.2 | 40.9 | 73.4 |