List your company

Tool-augmented planning on MagicEval-Tool Context Inheritance

98.8Name AccuracyMagicAgent-32B
-1.24824.72650.776.674Feb 22, 2026

Evaluation results

MethodLinks
MagicAgent-32BScale Category=MagicAgent Series2026.0298.885.5
MagicAgent-30B-A3BScale Category=MagicAgent Series2026.0298.582
Kimi-K2-InstructScale Category=Ultra-Scale2026.0290.367.5
Qwen3-MAXScale Category=Ultra-Scale2026.0289.567
Qwen3-235B-A22B-InstructScale Category=Ultra-Scale2026.0288.863.8
DeepSeek-V3.1-nothinkScale Category=Ultra-Scale2026.0286.362
GLM-4.7Scale Category=Ultra-Scale2026.0285.566
Llama3.3-70B-InstructScale Category=Large-Scale2026.0275.548.8
ERNIE-4.5-21B-A3B-PTScale Category=Large-Scale2026.0274.543.3
GPT-5.2Scale Category=Ultra-Scale2026.027155.5
Qwen3-32B-nothinkScale Category=Large-Scale2026.027048
Qwen3-30B-A3B-InstructScale Category=Large-Scale2026.0246.629.2
GLM-4.7-FlashScale Category=Large-Scale2026.0234.521.5
Olmo-3.1-32B-InstructScale Category=Large-Scale2026.022.614