WolfBench — leaderboard
WolfBench evaluates model capability on agentic tasks from the linked upstream source with Average as the primary reported metric.
Metric: Average Score (%). Source: wolfbench.ai. Status: saturation imminent. 32 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 87 |
| 2 | GPT-5.6 Terra | 82 |
| 3 | GPT-5.5 | 79 |
| 4 | GPT-5.6 Luna | 79 |
| 5 | Claude Opus 4.7 | 75 |
| 6 | Claude Sonnet 5 | 74 |
| 7 | Gemini 3.5 Flash | 72 |
| 8 | GPT-5.4 | 71 |
| 9 | Claude Opus 4.6 | 71 |
| 10 | GLM-5.2 | 71 |
| 11 | Claude Fable 5 | 68 |
| 12 | Claude Sonnet 4.6 | 62 |
| 13 | Kimi K2.6 | 60 |
| 14 | Gemini 3.1 Pro (Preview) | 59 |
| 15 | Kimi K2.7 Code | 58 |
Interactive version: theaggregate.ai/benchmark?slug=wolfbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.