safety benchmark
Siren AgentDojo evaluates tool-using agents under adversarial prompt-injection attacks using attack success rate.
Updated Sep 5, 2026
Higher score ranks better on this benchmark.
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelME | Score28.40% | Percentile100.00% | Participants1 | EvidenceC | Evaluated |
The leading models and scores on this benchmark.
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis siren agentdojo attack success rate AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
Selection summary
Muse Glimmer-30B currently leads Siren AgentDojo Attack Success Rate with 28.40%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price, and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What Siren AgentDojo Attack Success Rate measures and how its scores work.
Siren AgentDojo is a safety benchmark for evaluating tool-using agents under adversarial prompt-injection attacks.
It measures attack success rate as a ratio; lower is better.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
Common questions about Siren AgentDojo Attack Success Rate.
Muse Glimmer-30B is currently ranked first with 28.40%.
The current leaders are Muse Glimmer-30B (28.40%).
No matched official input price is currently available.
No matched runtime record is currently available.
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.
It measures attack success rate as a ratio; lower is better.
Yes. Higher values rank better for this benchmark.
1 model results are currently shown.
No. This benchmark is shown for reference but does not contribute to the overall score.