# Average Time Cost on MT-Bench benign prompts

Source: [Average Time Cost on MT-Bench benign prompts](<https://www.sota2.com/research/sota/average-time-cost-on-mt-bench-benign-prompts>)

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

Data updated: 2026-02-26T05:35:58.194Z

Evaluation focusing on the average inference time cost when running models against the benign prompt subset of the MT-Bench dataset.

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 | Large Language Model Evaluation |
| Dataset (including any stated split) | MT-Bench benign prompts |
| Metric | Average Time Cost |
| Unit | TIME |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [Average Time Cost](<https://www.sota2.com/research/sota/average-time-cost-on-mt-bench-benign-prompts.md?metric=Average%20Time%20Cost>): Lower is better; unit: TIME.

## Recorded results

| Method | Average Time Cost | Result date | Report paper |
| --- | --- | --- | --- |
| No defense | 41.56 | Not recorded | 68fa738f1e209131cea5d3a7 |
| Paraphrase | 45.42 | Not recorded | 68fa738f1e209131cea5d3a7 |
| Backtranslation w/ early termination | 49.86 | Not recorded | 68fa738f1e209131cea5d3a7 |
| Response Check | 65.84 | Not recorded | 68fa738f1e209131cea5d3a7 |
| Backtranslation w/o early termination | 86.91 | Not recorded | 68fa738f1e209131cea5d3a7 |
| SmoothLLM | 128.12 | Not recorded | 68fa738f1e209131cea5d3a7 |

### No defense

Average Time Cost: 41.56. Result ID: 699300653a6c31484912f75f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

Recorded evaluation configuration: {"Target Model":"Vicuna-13B"}. This record does not assert independent reproduction.

### Paraphrase

Average Time Cost: 45.42. Result ID: 699300653a6c31484912f761.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa707f1e209131cea4b7c6.md>)

Recorded evaluation configuration: {"Target Model":"Vicuna-13B"}. This record does not assert independent reproduction.

### Backtranslation w/ early termination

Average Time Cost: 49.86. Result ID: 699300653a6c31484912f764.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

[Implementation](<https://github.com/YihanWang617/LLM-Jailbreaking-Defense-Backtranslation>)

Recorded evaluation configuration: {"Target Model":"Vicuna-13B","early termination":true}. This record does not assert independent reproduction.

### Response Check

Average Time Cost: 65.84. Result ID: 699300653a6c31484912f762.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

Recorded evaluation configuration: {"Target Model":"Vicuna-13B"}. This record does not assert independent reproduction.

### Backtranslation w/o early termination

Average Time Cost: 86.91. Result ID: 699300653a6c31484912f763.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

[Implementation](<https://github.com/YihanWang617/LLM-Jailbreaking-Defense-Backtranslation>)

Recorded evaluation configuration: {"Target Model":"Vicuna-13B","early termination":false}. This record does not assert independent reproduction.

### SmoothLLM

Average Time Cost: 128.12. Result ID: 699300653a6c31484912f760.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d3a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73961e209131cea5d658.md>)

Recorded evaluation configuration: {"Target Model":"Vicuna-13B"}. This record does not assert independent reproduction.

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