CLBench - Procedural Task Execution: leaderboard
Metric: Solving Rate (%). Source: www.clbench.com. 36 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (xHigh) | 30.1 |
| 2 | Claude Opus 4.6 | 24.4 |
| 3 | GPT-5.1 (High) | 23.8 |
| 4 | Gemini 3.1 Pro (Preview) (High) | 23.8 |
| 5 | Grok 4.20 (Reasoning) | 23.4 |
| 6 | GPT-5.1 | 22.8 |
| 7 | Claude Opus 4.5 (Thinking) | 22.6 |
| 8 | Hy3-preview | 22.3 |
| 9 | GLM-5 | 21.6 |
| 10 | GPT-5.2 (High) | 21.4 |
| 11 | Seed 2.0 Pro (High) | 21.4 |
| 12 | Qwen 3.6 Plus | 21.3 |
| 13 | Kimi K2.5 | 19.5 |
| 14 | Claude Opus 4.5 | 19.5 |
| 15 | O3 (High) | 19.5 |
Interactive version: theaggregate.ai/benchmark?slug=clbench-procedural-task-execution · How It Works · Data refreshed daily, snapshot 2026-09-19.