Muse Spark: how it does on security work, and what it will answer

What happened

Muse Spark scores 65.4% end-to-end on Cybench on all 40 tasks, the highest full-task-set card figure on the leaderboard read 11 September 2026.

Cybench, unguided, solved: 65.4% (vendor-card; vendor system card run; 40 of 40 tasks; safeguards: as run by the vendor; safeguard state per the card; run 2026-09-11; CONFIRMED). Task count is what the vendor ran, so rows are comparable only with it beside them. Refusal gate (CONFIRMED, observed 2026-09-11): The Meta Model API Acceptable Use Policy bars creating malicious code, exploiting vulnerabilities without authorisation and building tools for offensive cyber operations; no verification tier.