Meta curated CTF set
Muse Spark 1.1 89.9% Agents solve 198 public and private capture-the-flag challenges across six security domains. Reports pass@1.
In-house Meta mix of public and private CTF challenges, first described in the Muse Spark Safety and Preparedness Report.
- Made by
- Meta
- Released
- April 2026
- Kind
- Capture-the-flag challenges
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- Muse Spark 1.1 #47 on the capability index
- Last in a system card
- July 2026
- Last model added
- July 2026
Muse Spark 1.1 leads with 89.9%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
- 1 Muse Spark 1.1 89.9%
- 2 Muse Spark 72%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Muse Spark 1.1 | 89.9% | Jul 2026 | Lab model card | |
| 2 | Muse Spark | 72% | Jun 2026 | Lab model card |
Meta’s own number for Muse Spark 1.1 passed 85% in July 2026
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Labs’ model cards
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
2 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Jul 9, 2026 | Muse Spark 1.1 system card | 89.9 | Curated CTFs pass@1, 198 challenges (Table 1, Table 4). pass@5 = 95.7 | Cyber risk |
| Jun 12, 2026 | Muse Spark system card | 72 | Curated CTF set pass@1, 198 challenges (Table 8, Figure 18). pass@10 86.4 | Cyber risk |
More in Cybersecurity
- ExploitGym In use 869 real-vulnerability exploit-development tasks (userspace, V8, Linux kernel). Success means capturing a flag through the intended vulnerability. GPT-6 Astra 42.4%
- ExploitBench In use 41 recent V8 JavaScript-engine N-day vulnerabilities. Progress measured by 16 grader-verified capability flags across five tiers. GPT-6 Astra 100%
- CyberBench v1.1 (Vals) In use Agents work on OSS-Fuzz memory-safety bugs from ARVO images in two tracks. They write a proof-of-concept input that crashes the target, or patch the source so the crash stops and tests still pass. Muse Wonder 80.4%
- SRE-Bench In use Software reverse engineering without source: 262 binary instances from 19 private programs, each with six graded objectives. GPT-6 Astra 56.9%
- CyScenarioBench In use Scenario-based long-horizon security challenges in sandboxes, organized in task, scenario and campaign tiers. Evaluation set kept private.
Gemini 4 Argon 100% - ExploitBench - Internal Port In use Agents develop working exploits for recently disclosed vulnerabilities to limit contamination. Scored as arbitrary code execution success rate. GPT-6 Cyber 42%