CESBench - Evaluation: leaderboard
The evaluation sub-domain of CESBench (Zhou et al., 2026): 51 items on module standards and security levels, leakage assessment and random number generator testing, across multiple-choice, judgment, scenario and code tasks. Score in percent, the mean of the four task-type scores within the sub-domain; higher is better.
Metric: Composite score (%; mean over the four task types of the item-score means within the D5 evaluation and certification sub-domain (51 items); 380 expert-written items on cryptographic engineering security for IoT devices; zero-shot, temperature 0, no tools, provider default reasoning mode; judgment and scenario answers scored by a Qwen3.5-397B judge from outside the evaluated set). Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi K2.6 | 80.5 | #173 |
| 2 | Gemini 3.7 Flash | 79.7 | #32 |
| 3 | GPT-5.6 Luna | 77.5 | #117 |
| 4 | DeepSeek V4 Pro | 77.3 | #181 |
| 5 | MiniMax-M2.5 | 73.6 | #590 |
| 6 | GLM-5.2 | 71.8 | #137 |
| 7 | DeepSeek V3 | 68.3 | #705 |
| 8 | Llama 4 Maverick | 60.2 | #1105 |
| 9 | GLM-4 32B (0414) | 59.4 | #1407 |
| 10 | Ling-flash-2.0 | 58.8 | #1352 |
| 11 | Hunyuan A13B-Instruct | 53.4 | #2021 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=cesbench-evaluation · How It Works · Data refreshed daily, snapshot 2026-11-12.