HWE-Bench v1.1 (RTL bug repair)
Version 1.1
Agents fix hardware bugs in 417 repository tasks. The revised tests check whether each fix restores the required behavior.
The authors changed 208 tasks, including hidden tests for 188 tasks. They evaluated seven agents on the revised release.
- Made by
- Peking University
- Released
- November 2026
- Kind
- Electronics
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Real work and use
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- HWE-Bench v1.0 (RTL bug repair)
- In views
- Agents
- Strongest model tested
- GPT-6 Astra #4 on the capability index
- Last model added
- October 2026
- Scores refreshed
- November 20th, 2026
Claude Opus 5.5 leads with 99%
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.
- 1 Claude Opus 5.5 max 99%
- 2 GPT-6 Astra max 97.1%
- 3 GPT-6.1 Sol max 97.1%
- 4 GPT-6 Sol max 93.5%
- 5
DeepSeek V4.1 Flash 1017 max 92.6%
- 6 Kimi K3.1 Pro max 92.6%
- 7 GLM-5.3 90.4%
Source: Official HWE-Bench v1.1 (RTL bug repair) leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 99% | max, Claude Code | Sep 2026 | Official leaderboard |
| 2 | GPT-6 Astra | 97.1% | max, Codex CLI | Sep 2026 | Official leaderboard |
| 3 | GPT-6.1 Sol | 97.1% | max, Codex CLI | Sep 2026 | Official leaderboard |
| 4 | GPT-6 Sol | 93.5% | max, Codex CLI | Sep 2026 | Official leaderboard |
| 5 | | 92.6% | max, DeepSeek Harness | Oct 2026 | Official leaderboard |
| 6 | Kimi K3.1 Pro | 92.6% | max, Kimi Code | Oct 2026 | Official leaderboard |
| 7 | GLM-5.3 | 90.4% | OpenCode | Aug 2026 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as OpenTitan, Ibex and CVA6. Claude Opus 5.5 leads 4 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
OpenTitan
7 models we track have a score here.
- 1 GPT-6 Astra ↑1 98.8%
- 2 Claude Opus 5.5 ↓1 98.4%
- 3 GPT-6.1 Sol 98.4%
- 4
DeepSeek V4.1 Flash 1017 ↑1 94.3%
- 5 Kimi K3.1 Pro ↑1 94.3%
- 6 GPT-6 Sol ↓2 93.5%
- 7 GLM-5.3 92.7%
Ibex
7 models we track have a score here.
- 1 Claude Opus 5.5 100%
- 2 GPT-6 Astra 97.1%
- 3 GPT-6 Sol ↑1 97.1%
- 4 GPT-6.1 Sol ↓1 97.1%
- 5 Kimi K3.1 Pro ↑1 88.6%
- 6
DeepSeek V4.1 Flash 1017 ↓1 85.7%
- 7 GLM-5.3 77.1%
CVA6
7 models we track have a score here.
- 1 Claude Opus 5.5 100%
- 2 GPT-6 Astra 97.1%
- 3 GPT-6 Sol ↑1 97.1%
- 4 GPT-6.1 Sol ↓1 97.1%
- 5
DeepSeek V4.1 Flash 1017 97.1%
- 6 GLM-5.3 ↑1 94.3%
- 7 Kimi K3.1 Pro ↓1 94.3%
Caliptra
7 models we track have a score here.
- 1 GLM-5.3 ↑6 100%
- 2 GPT-6 Astra 100%
- 3 Claude Opus 5.5 ↓2 100%
- 4 GPT-6 Sol 100%
- 5 GPT-6.1 Sol ↓2 100%
- 6
DeepSeek V4.1 Flash 1017 ↓1 100%
- 7 Kimi K3.1 Pro ↓1 100%
XiangShan
7 models we track have a score here.
- 1 Claude Opus 5.5 100%
- 2 GPT-6 Astra 96.3%
- 3 GPT-6.1 Sol 96.3%
- 4 GPT-6 Sol 92.6%
- 5
DeepSeek V4.1 Flash 1017 87%
- 6 Kimi K3.1 Pro 87%
- 7 GLM-5.3 83.3%
Rocket Chip
7 models we track have a score here.
- 1 Claude Opus 5.5 100%
- 2 GLM-5.3 ↑5 90.6%
- 3 GPT-6.1 Sol 87.5%
- 4
DeepSeek V4.1 Flash 1017 ↑1 87.5%
- 5 Kimi K3.1 Pro ↑1 87.5%
- 6 GPT-6 Astra ↓4 84.4%
- 7 GPT-6 Sol ↓3 84.4%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official HWE-Bench v1.1 (RTL bug repair) leaderboard
GLM-5.3 on the official leaderboard passed 85% in August 2026
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official HWE-Bench v1.1 (RTL bug repair) leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
Source: Official leaderboards
More in Engineering and CAD
- EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge.
Gemini 4 Carbon 83.9%
- 3DCodeBench In use Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity.
LongCat-2.5 55.4%
- CADGenBench Only in system cards Agents construct CAD models from design prompts. The generation split is graded for executability and geometric correctness. Claude Fable 5 45.1%
- PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%
- Razavi-Bench Not in model cards Models explain what happens in analog circuits and why. A language model awards up to four points per answer against a reference and a rubric.
Gemini 4 Carbon 98.3%
- CircuitRubric Not in model cards Models write analog circuit netlists from design requests. A program compares transistor connections and relative device sizes with accepted reference circuits. Claude Opus 4.8 96%
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (7 entries).
- Pass Models scored 7 models we track.
- Pass Newest score The newest is from October 2026, for Kimi K3.1 Pro.
- Pass Frontier models 3 of the top 10 have a score, led by GPT-6 Astra (#4).
- Note Headroom Saturated. The best is 99%, past 85% since August 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on pku-liang.github.io: 7 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.