Tool-augmented planning on MagicEval-Tool General
97.7Name AccuracyMagicAgent-30B-A3B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MagicAgent-30B-A3BScale Category=MagicAgent Series2026.02 | 97.7 | 86.3 | |
| MagicAgent-32BScale Category=MagicAgent Series2026.02 | 97.5 | 87.3 | |
| Qwen3-MAXScale Category=Ultra-Scale2026.02 | 92.4 | 72.3 | |
| Kimi-K2-InstructScale Category=Ultra-Scale2026.02 | 89.9 | 66.8 | |
| Qwen3-235B-A22B-InstructScale Category=Ultra-Scale2026.02 | 89.3 | 67.3 | |
| GLM-4.7Scale Category=Ultra-Scale2026.02 | 89.1 | 70.6 | |
| DeepSeek-V3.1-nothinkScale Category=Ultra-Scale2026.02 | 86.8 | 65 | |
| Llama3.3-70B-InstructScale Category=Large-Scale2026.02 | 74.9 | 53.8 | |
| GPT-5.2Scale Category=Ultra-Scale2026.02 | 74.6 | 55.8 | |
| ERNIE-4.5-21B-A3B-PTScale Category=Large-Scale2026.02 | 71.6 | 43.7 | |
| Qwen3-32B-nothinkScale Category=Large-Scale2026.02 | 69.5 | 50.7 | |
| Qwen3-30B-A3B-InstructScale Category=Large-Scale2026.02 | 48.3 | 34.5 | |
| GLM-4.7-FlashScale Category=Large-Scale2026.02 | 38.3 | 26.4 | |
| Olmo-3.1-32B-InstructScale Category=Large-Scale2026.02 | 19.5 | 11.7 |