# Hardware Verification Testbench Generation on CVDP-ECov

Source: [Hardware Verification Testbench Generation on CVDP-ECov](<https://www.sota2.com/research/sota/hardware-verification-testbench-generation-on-cvdp-ecov>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:40:05.132Z

Evaluation of model performance on generating hardware verification testbenches using the CVDP-ECov dataset, measured by coverage pass rates and average coverage scores.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Hardware Verification Testbench Generation |
| Dataset (including any stated split) | CVDP-ECov |
| Metric | Coverage Pass Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 28 of 28 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Coverage Pass Rate](<https://www.sota2.com/research/sota/hardware-verification-testbench-generation-on-cvdp-ecov.md?metric=Coverage%20Pass%20Rate>): Higher is better; unit: PERCENT.

- [Average Coverage Score](<https://www.sota2.com/research/sota/hardware-verification-testbench-generation-on-cvdp-ecov.md?metric=Average%20Coverage%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Coverage Pass Rate | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen3-235b-A22b-2507 | 69.9 | Not recorded | 6997eab209c33900fcd76730 |
| LLM4Cov | 69.2 | Not recorded | 6997eab209c33900fcd76730 |
| LLM4Cov | 67 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-Coder-30B-A3B | 63.9 | Not recorded | 6997eab209c33900fcd76730 |
| llama-4-maverick (400B) | 60.2 | Not recorded | 6997eab209c33900fcd76730 |
| LLM4Cov | 60.2 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen2.5-Coder-32B | 59.3 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-30B-A3B-2507 | 50.6 | Not recorded | 6997eab209c33900fcd76730 |
| LLM4Cov | 50.6 | Not recorded | 6997eab209c33900fcd76730 |
| LLM4Cov | 48.2 | Not recorded | 6997eab209c33900fcd76730 |
| LLM4Cov | 45.5 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen2.5-72B | 42.9 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-30B-A3B-2507 | 30.1 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-235b-A22b-2507 | 29.9 | Not recorded | 6997eab209c33900fcd76730 |
| CodeV-R1-RL-Qwen-7B | 29.9 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-Coder-30B-A3B | 28.4 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-4B-2507 (Base) | 28.4 | Not recorded | 6997eab209c33900fcd76730 |
| VeriCoder-Qwen2.5-14B | 28.2 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen2.5-Coder-32B | 23.4 | Not recorded | 6997eab209c33900fcd76730 |
| llama-4-maverick (400B) | 23.1 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen2.5-Coder-7B | 20.5 | Not recorded | 6997eab209c33900fcd76730 |
| VeriCoder-Qwen2.5-14B | 17.1 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen2.5-72B | 16.4 | Not recorded | 6997eab209c33900fcd76730 |
| VeriReason-Qwen2.5-7B | 15.9 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen3-4B-2507 (Base) | 14.9 | Not recorded | 6997eab209c33900fcd76730 |
| CodeV-R1-RL-Qwen-7B | 14.5 | Not recorded | 6997eab209c33900fcd76730 |
| Qwen2.5-Coder-7B | 11.3 | Not recorded | 6997eab209c33900fcd76730 |
| VeriReason-Qwen2.5-7B | 8.9 | Not recorded | 6997eab209c33900fcd76730 |

### Qwen3-235b-A22b-2507

Coverage Pass Rate: 69.9. Result ID: 699b01c75698665ec493c6f3.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### LLM4Cov

Coverage Pass Rate: 69.2. Result ID: 699b01c75698665ec493c70b.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"LLM4Cov","Training Stage":"Stage2"}. This record does not assert independent reproduction.

### LLM4Cov

Coverage Pass Rate: 67. Result ID: 699b01c75698665ec493c709.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"LLM4Cov","Training Stage":"Stage1"}. This record does not assert independent reproduction.

### Qwen3-Coder-30B-A3B

Coverage Pass Rate: 63.9. Result ID: 699b01c75698665ec493c6fb.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"Coding Specialized"}. This record does not assert independent reproduction.

### llama-4-maverick (400B)

Coverage Pass Rate: 60.2. Result ID: 699b01c75698665ec493c6f1.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### LLM4Cov

Coverage Pass Rate: 60.2. Result ID: 699b01c75698665ec493c707.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"LLM4Cov","Training Stage":"Stage0"}. This record does not assert independent reproduction.

### Qwen2.5-Coder-32B

Coverage Pass Rate: 59.3. Result ID: 699b01c75698665ec493c6f9.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b71e209131cea6981e.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"Coding Specialized"}. This record does not assert independent reproduction.

### Qwen3-30B-A3B-2507

Coverage Pass Rate: 50.6. Result ID: 699b01c75698665ec493c6f7.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### LLM4Cov

Coverage Pass Rate: 50.6. Result ID: 699b01c75698665ec493c70c.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"LLM4Cov","Training Stage":"Stage2"}. This record does not assert independent reproduction.

### LLM4Cov

Coverage Pass Rate: 48.2. Result ID: 699b01c75698665ec493c70a.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"LLM4Cov","Training Stage":"Stage1"}. This record does not assert independent reproduction.

### LLM4Cov

Coverage Pass Rate: 45.5. Result ID: 699b01c75698665ec493c708.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"LLM4Cov","Training Stage":"Stage0"}. This record does not assert independent reproduction.

### Qwen2.5-72B

Coverage Pass Rate: 42.9. Result ID: 699b01c75698665ec493c6f5.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b71e209131cea6981e.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### Qwen3-30B-A3B-2507

Coverage Pass Rate: 30.1. Result ID: 699b01c75698665ec493c6f8.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### Qwen3-235b-A22b-2507

Coverage Pass Rate: 29.9. Result ID: 699b01c75698665ec493c6f4.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### CodeV-R1-RL-Qwen-7B

Coverage Pass Rate: 29.9. Result ID: 699b01c75698665ec493c701.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c0e1e209131cea8c29f.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"Hardware Specific"}. This record does not assert independent reproduction.

### Qwen3-Coder-30B-A3B

Coverage Pass Rate: 28.4. Result ID: 699b01c75698665ec493c6fc.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"Coding Specialized"}. This record does not assert independent reproduction.

### Qwen3-4B-2507 (Base)

Coverage Pass Rate: 28.4. Result ID: 699b01c75698665ec493c705.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"LLM4Cov","Training Stage":"Base"}. This record does not assert independent reproduction.

### VeriCoder-Qwen2.5-14B

Coverage Pass Rate: 28.2. Result ID: 699b01c75698665ec493c6ff.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7acd1e209131cea85316.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"Hardware Specific"}. This record does not assert independent reproduction.

### Qwen2.5-Coder-32B

Coverage Pass Rate: 23.4. Result ID: 699b01c75698665ec493c6fa.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b71e209131cea6981e.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"Coding Specialized"}. This record does not assert independent reproduction.

### llama-4-maverick (400B)

Coverage Pass Rate: 23.1. Result ID: 699b01c75698665ec493c6f2.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### Qwen2.5-Coder-7B

Coverage Pass Rate: 20.5. Result ID: 699b01c75698665ec493c6fd.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b71e209131cea6981e.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"Coding Specialized"}. This record does not assert independent reproduction.

### VeriCoder-Qwen2.5-14B

Coverage Pass Rate: 17.1. Result ID: 699b01c75698665ec493c700.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7acd1e209131cea85316.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"Hardware Specific"}. This record does not assert independent reproduction.

### Qwen2.5-72B

Coverage Pass Rate: 16.4. Result ID: 699b01c75698665ec493c6f6.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b71e209131cea6981e.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"General Purpose"}. This record does not assert independent reproduction.

### VeriReason-Qwen2.5-7B

Coverage Pass Rate: 15.9. Result ID: 699b01c75698665ec493c703.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78981e209131cea79306.md>)

Recorded evaluation configuration: {"Execution Mode":"Agentic","Model Type":"Hardware Specific"}. This record does not assert independent reproduction.

### Qwen3-4B-2507 (Base)

Coverage Pass Rate: 14.9. Result ID: 699b01c75698665ec493c706.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"LLM4Cov","Training Stage":"Base"}. This record does not assert independent reproduction.

### CodeV-R1-RL-Qwen-7B

Coverage Pass Rate: 14.5. Result ID: 699b01c75698665ec493c702.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c0e1e209131cea8c29f.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"Hardware Specific"}. This record does not assert independent reproduction.

### Qwen2.5-Coder-7B

Coverage Pass Rate: 11.3. Result ID: 699b01c75698665ec493c6fe.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b71e209131cea6981e.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"Coding Specialized"}. This record does not assert independent reproduction.

### VeriReason-Qwen2.5-7B

Coverage Pass Rate: 8.9. Result ID: 699b01c75698665ec493c704.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab209c33900fcd76730.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78981e209131cea79306.md>)

Recorded evaluation configuration: {"Execution Mode":"Direct Inference","Model Type":"Hardware Specific"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
