Meta Muse Glimmer 30B Card - Siren AgentDojo - Utility: leaderboard
AgentDojo tool-use tasks (949 injection scenarios) attacked by an adaptive LLM attacker (Claude Opus 4.6); utility is the success rate of the user's benign task under injection. Meta Muse Glimmer 30B model card (August 2026); the methodology report gives internal runs for all three models, Glimmer at high reasoning and Gemma 4 31B and Qwen3.6-27B in thinking mode; percent, higher is better.
Metric: Utility Under Attack (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Muse Glimmer 30B (High) | 94.2 | #191 (Muse Glimmer 30B) |
| 2 | Qwen 3.6 27B (Reasoning) | 92.7 | #182 (Qwen 3.6 27B) |
| 3 | Gemma 4 31B (Reasoning) | 90.8 | #212 (Gemma 4 31B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=meta-muse-glimmer-30b-card-siren-agentdojo-utility · How It Works · Data refreshed daily, snapshot 2026-11-05.