Qwen3.7 Launch - MRCR-v2 128k — leaderboard
Metric: Accuracy (%). Source: qwen.ai. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.7 Max | 90.4 |
| 2 | Qwen 3.6 Plus | 85.9 |
| 3 | Claude Opus 4.6 (Thinking) | 84 |
| 4 | DeepSeek V4 Pro (Reasoning, Max Effort) | 74.4 |
| 5 | Kimi K2.6 (Thinking) | 63.1 |
| 6 | GLM-5.1 (Thinking) | 62 |
Interactive version: theaggregate.ai/benchmark?slug=qwen3-7-launch-mrcr-v2-128k · How the rankings work · Data refreshed daily, snapshot 2026-07-22.