MacAgentBench (OpenClaw) - Multi-App: leaderboard
Metric: Pass@1 (%; share of the 140 Multi-App tasks solved on a single attempt, deterministic rule-based checks; the OpenClaw in-container agent harness with shell, AppleScript and its skill library, at most 50 steps). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 (OpenClaw) | 63.6 |
| 2 | GPT-5.4 (OpenClaw) | 45 |
| 3 | Gemini 3.1 Pro (OpenClaw) | 40 |
| 4 | Qwen3-VL-235B-A22B (OpenClaw) | 22.9 |
| 5 | Qwen3-VL-32B-Thinking (OpenClaw) | 5.7 |
| 6 | Qwen3-VL-8B-Thinking (OpenClaw) | 2.9 |
| 7 | InternVL3.5-14B (OpenClaw) | 0 |
| 8 | InternVL3.5-8B (OpenClaw) | 0 |
Interactive version: theaggregate.ai/benchmark?slug=macagentbench-openclaw-multi-app · How It Works · Data refreshed daily, snapshot 2026-09-26.