HWE-Bench v1.0 (RTL bug repair)
Version 1.0
Agents fix bugs in six open-source hardware projects. The benchmark reports the share of fixes that pass repository tests in simulation.
The original release contains 417 tasks. The authors retain the v1.0 dataset tag and scores for comparisons.
- Made by
- Peking University
- Released
- April 2026
- Kind
- Electronics
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal, A simulator
- Items from
- Real work and use
- Refresh
- New versions replace it
- Run by
- Anyone
- Strongest model tested
- GPT-5.6 Sol #14 on the capability index
- Last model added
- July 2026
- Scores refreshed
- November 20th, 2026
GPT-5.6 Sol leads with 79.9% among models released by July 2026
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. Every run is in the next chart.
The newest model with a score here came out in July 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 15 more
- 1 GPT-5.6 Sol max 79.9%
- 2 GPT-5.5 xhigh 76.5%
- 3
DeepSeek V4 Flash 0731 max 76.5%
- 4 Claude Opus 4.7 max 74.6%
- 5 GLM-5.2 max 73.6%
- 6 GPT-5.4 xhigh 70.7%
- 7 Claude Opus 4.6 high 68.1%
- 8 Kimi K2.6 66.9%
- 9 Claude Sonnet 4.6 high 66.7%
- 10 GLM-5.1 63.1%
- 11
DeepSeek-V4-Pro max 60.4%
- 12
DeepSeek-V4-Flash max 58.3%
- 13
Qwen 3.6 Plus 52.5%
- 14
DeepSeek-V3.2 52%
- 15 Kimi K2.5 47.7%
Source: Official HWE-Bench v1.0 (RTL bug repair) leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.6 Sol | 79.9% | max, Codex CLI | Jul 2026 | Official leaderboard |
| 2 | GPT-5.5 | 76.5% | xhigh, Codex CLI | Apr 2026 | Official leaderboard |
| 3 | | 76.5% | max, OpenHands SDK | Jul 2026 | Official leaderboard |
| 4 | Claude Opus 4.7 | 74.6% | max, Claude Code | Apr 2026 | Official leaderboard |
| 5 | GLM-5.2 | 73.6% | max, OpenHands SDK | Jun 2026 | Official leaderboard |
| 6 | GPT-5.4 | 70.7% | xhigh, Codex CLI | Mar 2026 | Official leaderboard |
| 7 | Claude Opus 4.6 | 68.1% | high, Claude Code | Feb 2026 | Official leaderboard |
| 8 | Kimi K2.6 | 66.9% | Kimi CLI | Apr 2026 | Official leaderboard |
| 9 | Claude Sonnet 4.6 | 66.7% | high, Claude Code | Feb 2026 | Official leaderboard |
| 10 | GLM-5.1 | 63.1% | OpenHands SDK | Apr 2026 | Official leaderboard |
| 11 | | 60.4% | max, Claude Code | Apr 2026 | Official leaderboard |
| 12 | | 58.3% | max, Claude Code | Apr 2026 | Official leaderboard |
| 13 | | 52.5% | OpenHands SDK | Mar 2026 | Official leaderboard |
| 14 | | 52% | OpenHands SDK | Dec 2025 | Official leaderboard |
| 15 | Kimi K2.5 | 47.7% | Kimi CLI | Jan 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. GPT-5.6 Sol leads 2 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
OpenTitan
15 models we track have a score here.
- 1 GPT-5.6 Sol 75.1%
- 2
DeepSeek V4 Flash 0731 ↑1 71.8%
- 3 GPT-5.5 ↓1 71%
- 4 Claude Opus 4.7 70.6%
- 5 GLM-5.2 70.2%
- 6 GPT-5.4 64.1%
- 7 Claude Opus 4.6 63.7%
- 8 Claude Sonnet 4.6 ↑1 62%
- 9 GLM-5.1 ↑1 62%
- 10 Kimi K2.6 ↓2 61.6%
Ibex
15 models we track have a score here.
- 1 GPT-5.4 ↑5 91.4%
- 2 GPT-5.6 Sol ↓1 91.4%
- 3 GPT-5.5 ↓1 88.6%
- 4 Claude Opus 4.7 85.7%
- 5 Claude Opus 4.6 ↑2 82.9%
- 6 Claude Sonnet 4.6 ↑3 82.9%
- 7 Kimi K2.6 ↑1 80%
- 8
DeepSeek V4 Flash 0731 ↓5 80%
- 9 GLM-5.2 ↓4 77.1%
- 10 GLM-5.1 74.3%
CVA6
15 models we track have a score here.
- 1 GPT-5.4 ↑5 97.1%
- 2 GPT-5.5 94.3%
- 3 GPT-5.6 Sol ↓2 94.3%
- 4
DeepSeek V4 Flash 0731 ↓1 91.4%
- 5 Claude Opus 4.7 ↓1 88.6%
- 6 GLM-5.2 ↓1 88.6%
- 7 Claude Opus 4.6 85.7%
- 8 Claude Sonnet 4.6 ↑1 85.7%
- 9
Qwen 3.6 Plus ↑4 85.7%
- 10 Kimi K2.6 ↓2 85.7%
Caliptra
15 models we track have a score here.
- 1 Claude Opus 4.7 ↑3 100%
- 2
DeepSeek V4 Flash 0731 ↑1 100%
- 3 GPT-5.4 ↑3 93.8%
- 4 GPT-5.5 ↓2 93.8%
- 5 GLM-5.2 93.8%
- 6 GPT-5.6 Sol ↓5 93.8%
- 7 Claude Opus 4.6 87.5%
- 8
DeepSeek-V4-Flash ↑4 87.5%
- 9
DeepSeek-V3.2 ↑5 81.3%
- 10 Claude Sonnet 4.6 ↓1 81.3%
XiangShan
15 models we track have a score here.
- 1 GPT-5.5 ↑1 79.6%
- 2 GPT-5.6 Sol ↓1 77.8%
- 3 GLM-5.2 ↑2 75.9%
- 4
DeepSeek V4 Flash 0731 ↓1 75.9%
- 5 Claude Opus 4.7 ↓1 74.1%
- 6 GPT-5.4 68.5%
- 7 Claude Opus 4.6 66.7%
- 8 Claude Sonnet 4.6 ↑1 66.7%
- 9 Kimi K2.6 ↓1 63%
- 10
DeepSeek-V4-Pro ↑1 59.3%
Rocket Chip
15 models we track have a score here.
- 1 GPT-5.6 Sol 84.4%
- 2
DeepSeek V4 Flash 0731 ↑1 81.3%
- 3 Kimi K2.6 ↑5 71.9%
- 4 GPT-5.5 ↓2 71.9%
- 5 Claude Opus 4.7 ↓1 65.6%
- 6 GLM-5.2 ↓1 65.6%
- 7 GPT-5.4 ↓1 62.5%
- 8 Claude Opus 4.6 ↓1 59.4%
- 9
DeepSeek-V4-Pro ↑2 59.4%
- 10 Claude Sonnet 4.6 ↓1 56.3%
The best score, 79.9%, costs $1,179 for the full run
Runs on the cost frontier score higher than any run that's cheaper. DeepSeek V4 Flash 0731 at max comes within 3.4 points of the best for $40 for the full run, 30 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official HWE-Bench v1.0 (RTL bug repair) leaderboard
| Model | Effort | Harness | Score | Cost of the full run | Output tokens |
|---|---|---|---|---|---|
| GPT-5.6 Sol | max | Codex CLI | 79.9% | $1,179 | |
| GPT-5.5 | xhigh | Codex CLI | 76.5% | $1,064 | |
| max | OpenHands SDK | 76.5% | $40 | ||
| Claude Opus 4.7 | max | Claude Code | 74.6% | $561 | |
| GLM-5.2 | max | OpenHands SDK | 73.6% | $998 | |
| GPT-5.4 | xhigh | Codex CLI | 70.7% | $755 | |
| Claude Opus 4.6 | high | Claude Code | 68.1% | $213 | |
| Kimi K2.6 | Kimi CLI | 66.9% | $240 | ||
| Claude Sonnet 4.6 | high | Claude Code | 66.7% | $108 | |
| GLM-5.1 | OpenHands SDK | 63.1% | $225 | ||
| max | Claude Code | 60.4% | $33 | ||
| max | Claude Code | 58.3% | $12 | ||
| OpenHands SDK | 52.5% | $209 | |||
| OpenHands SDK | 52% | $42 | |||
| Kimi K2.5 | Kimi CLI | 47.7% | $67 |
4 models have held number one on the official leaderboard since its release in April 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official HWE-Bench v1.0 (RTL bug repair) leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official HWE-Bench v1.0 (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 (15 entries).
- Pass Models scored 15 models we track.
- Pass Newest score The newest is from July 2026, for DeepSeek V4 Flash 0731.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5.6 Sol (#14).
- Pass Headroom The best is 79.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on pku-liang.github.io: 15 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.