An Empirical Study of Proactive Coding Assista — leaderboard
Metric: Pass@1 (self-reported). Source: benchmarklist.com. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 13.57 |
| 2 | Gemini 3.1 Pro (Preview) | 11.46 |
| 3 | Qwen 3.5 397B A17B | 8.43 |
| 4 | DeepSeek V3.2 | 7.77 |
| 5 | GLM-5 | 7.77 |
| 6 | GPT-5.4 | 6.59 |
| 7 | MiniMax-M2.5 | 2.77 |
Interactive version: theaggregate.ai/benchmark?slug=an-empirical-study-of-proactive-coding-assista · How the rankings work · Data refreshed daily, snapshot 2026-07-22.