Meta Muse Spark Open Card - OSWorld Verified (361 tasks, GUI, mean@4): leaderboard
Meta Muse Spark Open release evaluation: Computer control on 361 Ubuntu desktop tasks, GUI only, 200-step cap; mean task reward across four attempts. Muse and Inkling native coordinate conventions are retained. Only internal runs; Muse/Inkling xhigh and Kimi max where included.
Metric: Mean task reward (%). Source: huggingface.co. Saturation forecast: Around January 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Muse Spark Open (xHigh) | 80.4 | #22 (Muse Spark Open) |
| 2 | Inkling (xHigh) | 40.1 | #122 (Inkling) |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=meta-muse-spark-open-card-osworld-verified-361-tasks-gui-mean-4 · How It Works · Data refreshed daily, snapshot 2026-10-16.