LoopsBench - Vendor Loops - Dependency Depth: leaderboard
LoopsBench (arXiv 2608.00267v2, Table 2 within-family panel): 12 models on 112 dependency-structured long-horizon coding tasks, each run in its vendor's loop (GPT in Codex, Claude in Claude Code, Qwen in Qwen Code). Mean fraction of the task's prerequisite layers fully passing before the first residual handoff; 0-1, higher is better.
Metric: Dependency depth reached before the first residual handoff (0-1). Source: arxiv.org. Saturation forecast: Around February 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 0.61 | #51 |
| 2 | GPT-5.5 | 0.53 | #33 |
| 3 | Claude Opus 4.5 | 0.46 | #87 |
| 4 | GPT-5.2 | 0.36 | #115 |
| 5 | Claude Sonnet 4.5 | 0.34 | #148 |
| 6 | Qwen 3.6 Plus | 0.31 | #124 |
| 7 | GPT-5 | 0.22 | #105 |
| 8 | Qwen 3.5 Plus | 0.19 | #133 |
| 9 | Claude Sonnet 4 | 0.18 | #207 |
| 10 | GPT-4o | 0.09 | #357 |
| 11 | Qwen 3 Max | 0.07 | #213 |
| 12 | Qwen 2.5 72B | 0.03 | #512 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=loopsbench-vendor-loops-dependency-depth · How It Works · Data refreshed daily, snapshot 2026-11-05.