HierSVA-B - Proof Success: leaderboard
Metric: Assertion proof success rate (%, C2-P): share of generated assertions in evaluable runs that are proven non-vacuously on HierSVA-DS (342 hierarchical BaseJump STL modules, depths 0 to 9): static generation of SystemVerilog Assertions from RTL and submodule contracts, checked by formal proof in VC Formal; one run, reasoning enabled, temperature 0, strict JSON output. Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.6 Max (Preview) (Thinking) | 95.2 |
| 2 | GLM-5.1 | 91 |
| 3 | Gemini 3.1 Pro (Preview) | 90.3 |
| 4 | Claude Opus 4.7 (Thinking) | 87.5 |
| 5 | Kimi K2.6 | 86.2 |
| 6 | GPT-5 Mini | 77.9 |
| 7 | Qwen 3.6 Plus (Thinking) | 76.9 |
| 8 | GPT-5.5 | 76.8 |
| 9 | Claude Haiku 4.5 (Thinking) | 74 |
| 10 | MiniMax-M2.7 | 58.2 |
Interactive version: theaggregate.ai/benchmark?slug=hiersva-b-proof-success · How It Works · Data refreshed daily, snapshot 2026-09-29.