Charlemagne Labs social-engineering benchmark
Muse Spark 1.3 1.60% Attacker models try to persuade a skeptical simulated victim to take an action over many turns. Scored by attempted and carry-through rates.
Used in Meta Muse Spark safety reports. Filtered to 280 scenarios. Attempted and carry-through tracked separately.
- Made by
- Charlemagne Labs
- Kind
- Refusing harmful requests
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- Muse Spark 1.3 #26 on the capability index
- Last in a system card
- October 2026
- Last model added
- October 2026
Muse Spark 1.3 leads with 1.60%
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 1.3 1.60%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Muse Spark 1.3 | 1.60% | Oct 2026 | Lab model card |
Scores over time
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.
3 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 |
|---|---|---|---|---|
| Oct 23, 2026 | Muse Spark 1.3 system card | 1.6 | Social Engineering Propensity, attempted rate, Muse Spark 1.3 (API) (Figure 9). Carry-through rate 0.0 | Cyber risk |
| Jul 9, 2026 | Muse Spark 1.1 system card | 5.1 | Charlemagne Labs multi-turn social engineering composite score 0 to 100, non-refused prompts (Table 1, Figure 12) | Cyber risk |
| Jun 12, 2026 | Muse Spark system card | 57.3 | Charlemagne Labs multi-turn social engineering composite capability score 0 to 100 (Table 1, Figure 20). Deployed-system refusal rate 99.9 (Table 2) | Cyber risk |
More in Safety and alignment
- FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
- StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is.
Inkling 98.6% - Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts.
Gemini 4 Carbon 0% - DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics. Muse Spark 1.2 88.4%
- Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
- IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%