WildClawBench — leaderboard
WildClawBench evaluates model capability on agentic tasks from the linked upstream source with Overall Score as the primary reported metric.
Metric: Overall Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 62.2 |
| 2 | GPT-5.5 | 58.2 |
| 3 | Nex N2 Pro | 53.5 |
| 4 | Claude Opus 4.6 | 51.6 |
| 5 | GPT-5.4 | 50.3 |
| 6 | GLM-5.1 | 48.2 |
| 7 | DeepSeek V4 Pro | 43.7 |
| 8 | GLM-5 | 42.6 |
| 9 | Gemini 3.1 Pro (Preview) | 40.8 |
| 10 | MiMo-V2-Pro | 40.2 |
| 11 | Qwen 3.5 397B A17B | 34.5 |
| 12 | DeepSeek V3.2 | 34 |
| 13 | GLM-5-Turbo | 33.9 |
| 14 | MiniMax-M2.7 | 33.8 |
| 15 | MiMo-V2-Flash | 30.8 |
Interactive version: theaggregate.ai/benchmark?slug=wildclawbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.