SuperCLUE General (May 2026) - Agentic Task Planning: leaderboard
Metric: Score. Source: www.superclueai.com. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (High) | 86.56 |
| 2 | Qwen 3.6 Max (Preview) (Thinking) | 83.41 |
| 3 | Kimi K2.6 (Thinking) | 80.95 |
| 4 | DeepSeek V4 Pro (Max) | 78.12 |
| 5 | MiniMax-M3 | 75.91 |
| 6 | Claude Opus 4.7 (High) | 75.7 |
| 7 | DeepSeek V4 Flash (Max) | 75.56 |
| 8 | Gemini 3.1 Pro (Preview) (High) | 75.12 |
| 9 | Spark X2 | 71.83 |
| 10 | Claude Opus 4.8 (High) | 71.63 |
| 11 | Ernie 5.1 | 70.44 |
| 12 | Gemini 3.5 Flash (High) | 69.6 |
| 13 | Step 3.7 Flash | 67.24 |
| 14 | GLM-5.1 | 66.8 |
| 15 | Step 3.5 Flash | 64.99 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-may-2026-agentic-task-planning · How It Works · Data refreshed daily, snapshot 2026-09-19.