Boiling the Frog: leaderboard
Background. Traditional safety benchmarks for language models evaluate generated text: whether a model outputs toxic language, reproduces bias, or follows harmful instructions. When models are deployed as agents, the safety-relevant object shifts from what the system says to what it does within an environment, and evaluating model responses under prompting is no longer sufficient to address the sa
Metric: Strict ASR (self-reported). Source: benchmarklist.com. Saturation forecast: Around September 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Haiku 4.5 | 20.5 | #280 |
| 2 | GLM-5.1 | 20.6 | #97 |
| 3 | GPT-5.3 Codex | 23.1 | #69 |
| 4 | Kimi K2.6 | 25 | #104 |
| 5 | MiniMax-M2.7 | 37.8 | #258 |
| 6 | DeepSeek V4 Pro | 41.7 | #102 |
| 7 | Seed 2.0 Lite | 87.2 | #145 |
| 8 | Gemini 3.1 Flash Lite | 92.9 | #191 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=boiling-the-frog · How It Works · Data refreshed daily, snapshot 2026-10-20.