WhitzardNUWA Lab ↗
NUWANUWA INDEX

Search NUWA Index

Research · Benchmarks · Findings · Perspectives

中文
← Models

Meta

Muse Spark

Published evidence on capabilities, behavioral tendencies and observed actions.

First public availability
2026-04-08
Parameters
Not disclosed
Metadata updated
2026-10-27
Sources ↓

Evidence overview

Developer documentation is linked below. Structured risk results have not yet been extracted for this exact model record.

NUWA curates and interprets these sources. Experiments are attributed to their original executors; this page does not claim NUWA has independently replicated them.

MODEL MONITORING

Representative benchmark coverage

Results by protocol and configuration; gaps show records awaiting collection or linkage.

Cybench →

No eligible measurement linked to this model version.

MASK →

No eligible measurement linked to this model version.

All linked measurements

Each result retains its protocol and unit.

6 observations

Measurements and source conditions
System / modelResultCondition statusTask / conditionsEvaluator / reportDates
Muse Spark · Meta AgentMuse Spark43.5%CyberGym · Success rateIncompletelevel1score_10Meta AgentNot disclosedCyberGym official leaderboard and machine-data extractionRelease 2026-04-08Report —
Muse Spark47.7%Agentic misalignment · deliberate harmful action rateIncompleteAnthropic Agentic Misalignmentdeliberate harmful action rate—MetaEvaluating DeepSeek v4 Pro for Frontier RisksRelease 2026-04-08Report 2026-05-29
Muse Spark1.6%DeceptionBench · deception score: single-turnIncompleteDeceptionBenchdeception score: single-turn—MetaEvaluating DeepSeek v4 Pro for Frontier RisksRelease 2026-04-08Report 2026-05-29
Muse Spark1.9%DeceptionBench · deception score: multi-turn w pressureIncompleteDeceptionBenchdeception score: multi-turn w pressure—MetaEvaluating DeepSeek v4 Pro for Frontier RisksRelease 2026-04-08Report 2026-05-29
Muse Spark1.8 percentage pointsAlignment faking · Compliance gapIncompleteAlignment FakingCompliance gap—MetaEvaluating DeepSeek v4 Pro for Frontier RisksRelease 2026-04-08Report 2026-05-29
Muse Spark0%Alignment faking · Deceptive alignment rateIncompleteAlignment FakingDeceptive alignment rate—MetaEvaluating DeepSeek v4 Pro for Frontier RisksRelease 2026-04-08Report 2026-05-29

Developer disclosures

External assessments

CyberGym official snapshot →

Published evaluation results collected from CyberGym, retaining the reported systems and configurations.

Publication date not established

Evidence by risk topic

Autonomy 1 protocol results

Agentic Misalignment →

Reused result · Executed by Meta · Original executor Meta

deliberate harmful action rate
47.7%

Per-result sample size not disclosed

Deception 5 protocol results

DeceptionBench →

Reused result · Executed by Meta · Original executor Meta

deception score: single-turn
1.6%

Per-result sample size not disclosed

DeceptionBench →

Reused result · Executed by Meta · Original executor Meta

deception score: multi-turn w pressure
1.9%

Per-result sample size not disclosed

Alignment faking →

Reused result · Executed by Meta · Original executor Meta

Compliance gap
1.8 percentage points

Per-result sample size not disclosed

Alignment faking →

Reused result · Executed by Meta · Original executor Meta

Deceptive alignment rate
0%

Per-result sample size not disclosed

Agentic Misalignment →

Reused result · Executed by Meta · Original executor Meta

deliberate harmful action rate
47.7%

Per-result sample size not disclosed

Cybersecurity 1 protocol results
CBRN 1 protocol results

WMDP →

Reused result · Executed by Meta · Original executor Meta

accuracy
89.8%

Per-result sample size not disclosed

Disinformation No structured result linked

A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.

Autonomous physical harm No structured result linked

A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.

Source review and evidence gaps

CyberGym →

Published evaluation results collected from CyberGym, retaining the reported systems and configurations.

The site reports submitted systems under different configurations. Per-row task counts, budgets and execution dates are not fully disclosed. Undefined metrics and unresolved identity rows are excluded. Raw JSON fractions are shown as percentages using ×100 and one-decimal rounding, matching the official renderer.

Evaluating DeepSeek v4 Pro for Frontier Risks →

Source-specific extraction of Neo Research · DeepSeek V4 Pro. Findings and measurements retain their original attribution.

Source transcription is not experimental replication. Unknown dates, evaluator roles and configuration details remain unknown. Source groups are not verified rankings.

Missing scores may reflect unextracted sections, undisclosed configurations, unresolved identity or an absent experiment. These are separate research questions; none implies a zero result.

Interpretation and open questions

Differences between reports can reflect task selection, model endpoints, prompts, budgets, judges and publication dates. Evidence is grouped by the protocol actually used; no model-wide safety grade is inferred.

References

  1. [1] Meta. Meta — muse-spark. 2026-04-08 ; Dated launch and explicit availability; Parameter counts not disclosed in reviewed developer source
  2. [2] Evaluating DeepSeek v4 Pro for Frontier Risks. 2026-05-29. v1.1 · 29 May 2026 ; PDF p. 9; WMDP-Cyber; accuracy; PDF p. 27; DeceptionBench; deception score: single-turn; PDF p. 27; DeceptionBench; deception score: multi-turn w pressure
  3. [3] Muse Spark Safety & Preparedness Report (Meta-hosted May 26 revision). 2026-05-26
  4. [4] Muse Spark Safety & Preparedness Report. 2026-05-14. 2026-08-24
  5. [5] CyberGym official leaderboard and machine-data extraction ; /level1/49/score_10