SuperCLUE General (July 2026) - Agentic Task Planning: leaderboard

Metric: Score. Source: www.superclueai.com. 18 models tracked.

Top models

#ModelScore
1GLM-5.3 (Max)91.15
2Qwen 3.8 Max (Max)90.94
3Qwen 3.8 Flash (Max)88.61
4GLM-5.3 Flash (Max)88.43
5DeepSeek V4 Pro (0813) (Max)85.48
6Kimi K3 (Max)84.35
7DeepSeek V4 Flash (Max)80.72
8DeepSeek V4.1 Flash (Max)79.84
9DeepSeek V4 Pro (Max)79
10MiniMax-M377.68
11LongCat 2.071.32
12Hy3 (High)66.82
13Step 3.7 Flash (High)66.54
14GLM-5.2 (Max)65.27
15MiMo-V2.5-Pro57.25

Interactive version: theaggregate.ai/benchmark?slug=superclue-general-july-2026-agentic-task-planning · How It Works · Data refreshed daily, snapshot 2026-09-19.