MalOptBench: leaderboard
Metric: Attack Success Rate (%). Source: arxiv.org. Saturation forecast: Around February 2027. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 38.33 |
| 2 | Qwen 3 235B A22B 2507 Instruct | 51.67 |
| 3 | O3 (2025-04-16) | 55 |
| 4 | Phi-4 | 65 |
| 5 | Claude Sonnet 4 (20250514) | 81.66 |
| 6 | GPT-4o (2024-08-06) | 96.66 |
| 7 | Command A | 98.33 |
| 8 | Gemini 2.5 Flash | 100 |
| 9 | DeepSeek V3 | 100 |
| 10 | DeepSeek V3.1 | 100 |
| 11 | Grok 3 Mini | 100 |
Interactive version: theaggregate.ai/benchmark?slug=maloptbench · How It Works · Data refreshed daily, snapshot 2026-09-25.