SuperCLUE General (July 2026) - Agentic Task Planning: leaderboard
SuperCLUE general leaderboard, July 2026 edition (2026年7月): agentic task planning (智能体任务规划) score from the 总排行榜 sheet. SuperCLUE's periodic Chinese-language general evaluation uses a new question set each edition, so editions are separate boards. Score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Around February 2027. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.3 (Max) | 91.15 | #44 (GLM-5.3) |
| 2 | Qwen 3.8 Max (Max) | 90.94 | #34 (Qwen 3.8 Max) |
| 3 | Qwen 3.8 Flash (Max) | 88.61 | #62 (Qwen 3.8 Flash) |
| 4 | GLM-5.3 Flash (Max) | 88.43 | #77 (GLM-5.3 Flash) |
| 5 | DeepSeek V4 Pro (0813) (Max) | 85.48 | #58 (DeepSeek V4 Pro (0813)) |
| 6 | Doubao-Seed-2.1-Pro (High) | 84.95 | |
| 7 | Kimi K3 (Max) | 84.35 | #35 (Kimi K3) |
| 8 | DeepSeek V4 Flash (Max) | 80.72 | #155 (DeepSeek V4 Flash) |
| 9 | DeepSeek V4.1 Flash (Max) | 79.84 | #37 (DeepSeek V4.1 Flash) |
| 10 | DeepSeek V4 Pro (Max) | 79 | #112 (DeepSeek V4 Pro) |
| 11 | MiniMax-M3 | 77.68 | #151 |
| 12 | LongCat 2.0 | 71.32 | #220 |
| 13 | Hy3 (High) | 66.82 | #100 (Hy3) |
| 14 | Step 3.7 Flash (High) | 66.54 | #247 (Step 3.7 Flash) |
| 15 | GLM-5.2 (Max) | 65.27 | #80 (GLM-5.2) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-july-2026-agentic-task-planning · How It Works · Data refreshed daily, snapshot 2026-11-05.