ClashBench - Deliberate Interference Rate (OpenCode): leaderboard
Metric: DIR (%). Source: arxiv.org. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.7 Flash | 68.6 |
| 2 | Qwen 3.7 Plus | 65.6 |
| 3 | Qwen 3.8 27B | 64.9 |
| 4 | Qwen 3.8 Max | 63.5 |
| 5 | Qwen 3.6 27B | 61.5 |
| 6 | GLM-5.3 | 60.5 |
| 7 | Qwen 3.5 9B | 57.1 |
| 8 | Qwen 3.6 35B A3B | 56.9 |
| 9 | DeepSeek V4 Flash (0731) | 55.7 |
| 10 | GLM-5.2 | 52.4 |
| 11 | GLM-4.7 | 50.2 |
| 12 | MiniMax-M3 | 46.1 |
Interactive version: theaggregate.ai/benchmark?slug=clashbench-deliberate-interference-rate-opencode · How It Works · Data refreshed daily, snapshot 2026-09-20.