HWE-Bench (OpenHands SDK): leaderboard
HWE-bench (PKU) run with the OpenHands SDK agent scaffold: 417 simulation-verified RTL bug-repair cases across six open hardware projects; percent of cases resolved.
Metric: Resolved (%). Source: pku-liang.github.io. Saturation forecast: Around February 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Flash (0731) (Max) | 76.5 | #84 (DeepSeek V4 Flash (0731)) |
| 2 | GLM-5.2 (Max) | 73.6 | #80 (GLM-5.2) |
| 3 | GLM-5.1 | 63.1 | #109 |
| 4 | Qwen 3.6 Plus | 52.5 | #124 |
| 5 | DeepSeek V3.2 | 52 | #210 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=hwe-bench-openhands-sdk · How It Works · Data refreshed daily, snapshot 2026-11-05.