VEHBench - Design Search: leaderboard
VEHBench (2026) is a stage-local diagnostic benchmark for LLM-assisted design of vibration energy harvesters for battery-free IoT, scored against a physics oracle. Mean final feasible power relative to the best known feasible design after oracle-guided edits, a ratio where 1 matches the best known design; higher is better.
Metric: P2 final feasible power ratio (ratio, 0 and up, 1 = best known feasible design): mean over tasks of final feasibility times output power relative to the best known feasible power, infeasible final designs scoring zero, after oracle-feedback-guided bounded edits; LLM-assisted vibration energy harvester design against a physics oracle; one complete run per model. Source: arxiv.org. Saturation forecast: Around September 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 0.39 | #94 |
| 2 | DeepSeek R1 | 0.24 | #473 |
| 3 | Claude Sonnet 4.6 | 0.24 | #161 |
| 4 | Qwen 3.6 Plus | 0.21 | #222 |
| 5 | Qwen 3 Max | 0.2 | #403 |
| 6 | Hy3-preview (Reasoning) | 0.16 | #316 (Hy3-preview) |
| 7 | DeepSeek V3 | 0.16 | #716 |
| 8 | O4 Mini | 0.16 | #353 |
| 9 | GPT-5.4 | 0.13 | #123 |
| 10 | DeepSeek V4 Pro (Non-reasoning) | 0.13 | #188 (DeepSeek V4 Pro) |
| 11 | Llama 3.3 70B | 0.12 | #1498 |
| 12 | MiMo-V2.5-Pro (Non-reasoning) | 0.11 | #279 (MiMo-V2.5-Pro) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=vehbench-design-search · How It Works · Data refreshed daily, snapshot 2026-11-13.