RoadmapBench: leaderboard
Metric: Resolved (%) (self-reported). Source: benchmarklist.com. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 39.1 |
| 2 | Claude Opus 4.6 | 32.2 |
| 3 | GPT-5.4 | 29.6 |
| 4 | Gemini 3.1 Pro (Preview) | 20.9 |
| 5 | GLM-5.1 | 20.9 |
| 6 | DeepSeek V4 Pro | 18.3 |
| 7 | Qwen 3.6 Plus | 16.5 |
| 8 | Kimi K2.6 | 15.7 |
| 9 | MiMo-V2.5-Pro | 13.9 |
| 10 | Kimi K2.5 | 11.3 |
| 11 | Qwen 3.5 397B A17B | 10.4 |
| 12 | MiniMax-M2.7 | 10.4 |
| 13 | Seed 2.0 Pro | 5.2 |
Interactive version: theaggregate.ai/benchmark?slug=roadmapbench · How It Works · Data refreshed daily, snapshot 2026-09-05.