CESBench - Integration: leaderboard
The integration sub-domain of CESBench (Zhou et al., 2026): 53 items on key management, trusted execution, secure boot, firmware update and public-key infrastructure, 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 D6 integration and key management sub-domain (53 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: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.2 | 90.7 | #137 |
| 2 | Kimi K2.6 | 89.7 | #173 |
| 3 | DeepSeek V4 Pro | 87.6 | #181 |
| 4 | MiniMax-M2.5 | 86.5 | #590 |
| 5 | GPT-5.6 Luna | 84.3 | #117 |
| 6 | Gemini 3.7 Flash | 81.5 | #32 |
| 7 | DeepSeek V3 | 79.3 | #705 |
| 8 | Llama 4 Maverick | 73.8 | #1105 |
| 9 | Ling-flash-2.0 | 73.5 | #1352 |
| 10 | GLM-4 32B (0414) | 68.9 | #1407 |
| 11 | Hunyuan A13B-Instruct | 64.9 | #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-integration · How It Works · Data refreshed daily, snapshot 2026-11-12.