PrepBench - GUI Workflow: leaderboard
Metric: End-to-end GUI workflow accuracy (%): from the original ambiguous request, the agent must produce an executable operator workflow whose output table matches the reference, PrepBench natural-language data-preparation tasks, LLM agent at temperature 0.7 with Clarify (budgeted, answered by a DeepSeek-V3.2 user simulator), Profile, Code and Translate actions; higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.1 Codex | 34.6 |
| 2 | Kimi K2 (Thinking) | 30.1 |
| 3 | Claude Sonnet 4.5 | 24.5 |
| 4 | Gemini 3 Flash | 22.2 |
| 5 | GLM-4.7 | 19.9 |
| 6 | Qwen 3 235B A22B | 19.3 |
| 7 | DeepSeek V3.2 | 15.7 |
| 8 | Grok Code Fast 1 | 13.1 |
| 9 | Devstral 2 | 8.5 |
| 10 | GPT-4o | 5.2 |
Interactive version: theaggregate.ai/benchmark?slug=prepbench-gui-workflow · How It Works · Data refreshed daily, snapshot 2026-10-07.