QwenClawBench: leaderboard
Real-user-distribution Claw agent benchmark referenced in Qwen's Qwen3.7-Max launch post.
Metric: Score (%). Source: github.com. Source: benchmarklist.com. Saturation forecast: Around March 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 (Max) | 65.5 | #63 (Claude Opus 4.6) |
| 2 | Qwen 3.7 Plus | 61.8 | #92 |
| 3 | DeepSeek V4 Pro (Max) | 59.2 | #102 (DeepSeek V4 Pro) |
| 4 | GLM-5.1 (Thinking) | 58.7 | #97 (GLM-5.1) |
| 5 | Qwen 3.6 Plus | 57.2 | #120 |
| 6 | Kimi K2.6 (Thinking) | 54.7 | #104 (Kimi K2.6) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=qwenclawbench · How It Works · Data refreshed daily, snapshot 2026-10-20.