Tool-augmented planning on MagicEval-Tool Condition
95.4Name AccuracyMagicAgent-32B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MagicAgent-32BScale Category=MagicAgent Series2026.02 | 95.4 | 88.7 | |
| MagicAgent-30B-A3BScale Category=MagicAgent Series2026.02 | 94.2 | 87.2 | |
| Kimi-K2-InstructScale Category=Ultra-Scale2026.02 | 88.7 | 71 | |
| Qwen3-MAXScale Category=Ultra-Scale2026.02 | 88.1 | 69.5 | |
| Qwen3-235B-A22B-InstructScale Category=Ultra-Scale2026.02 | 87.5 | 69.2 | |
| DeepSeek-V3.1-nothinkScale Category=Ultra-Scale2026.02 | 86.6 | 66.5 | |
| GLM-4.7Scale Category=Ultra-Scale2026.02 | 82.6 | 65.6 | |
| Llama3.3-70B-InstructScale Category=Large-Scale2026.02 | 76.8 | 53.1 | |
| GPT-5.2Scale Category=Ultra-Scale2026.02 | 73.5 | 58.2 | |
| ERNIE-4.5-21B-A3B-PTScale Category=Large-Scale2026.02 | 61.3 | 40.2 | |
| Qwen3-32B-nothinkScale Category=Large-Scale2026.02 | 57.6 | 44.2 | |
| Qwen3-30B-A3B-InstructScale Category=Large-Scale2026.02 | 40.2 | 29.7 | |
| GLM-4.7-FlashScale Category=Large-Scale2026.02 | 27.4 | 20.1 | |
| Olmo-3.1-32B-InstructScale Category=Large-Scale2026.02 | 10.7 | 5.2 |