SuperCLUE-Agent (October 2023) - Task Planning: leaderboard
SuperCLUE-Agent evaluates Chinese LLM agent capabilities: task planning, tool use and long- and short-term memory. October 2023 edition (2023年10月): task planning total covering task decomposition, self-reflection and chain of thought. Each dated workbook is a separate evaluation round, kept apart from the current board. Score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 81.88 | #459 |
| 2 | GPT-3.5 Turbo | 61.07 | #858 |
| 3 | MiniMax-abab5.5 | 46.11 | #1030 |
| 4 | Baichuan2-13B-Chat | 40.48 | #1210 |
| 5 | Llama 2 13B Chat | 38.59 | #1414 |
| 6 | Qwen-7B Chat | 28.52 | #1353 |
| 7 | Chinese-Alpaca-2-13B | 23.65 | #1390 |
| 8 | ChatGLM2 6B | 19.09 | #1475 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-agent-october-2023-task-planning · How It Works · Data refreshed daily, snapshot 2026-11-05.