CoDyControlBench - SMC-Family Controllers: leaderboard
The sliding-mode-control half of CoDyControlBench (2026): percentage of simulated dynamical systems for which the model designs a sliding-mode-family controller that meets the steady-state and overshoot criteria under the Fixed-3 protocol. Higher is better.
Metric: Design success rate (%; sliding-mode-control family configurations under the Fixed-3 protocol; 132 controller-design configurations (1-6 DoF, LTI/LTV/nonlinear systems, coupling, damping, SMC- and PID-family controllers), mean of three independent runs; a design succeeds when every response channel meets the steady-state band and overshoot limits). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (High) | 94.9 | #49 (GPT-5.5) |
| 2 | Gemini 3.1 Pro (Preview) | 94.7 | #94 |
| 3 | GLM-5.2 | 89.6 | #138 |
| 4 | Claude Opus 4.8 | 75.8 | #56 |
| 5 | Qwen 3.5 397B A17B | 65.4 | #246 |
| 6 | DeepSeek V4 Pro (High) | 62.9 | #188 (DeepSeek V4 Pro) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=codycontrolbench-smc-family-controllers · How It Works · Data refreshed daily, snapshot 2026-11-13.