# Output Prediction on CruxEval

Source: [Output Prediction on CruxEval](<https://www.sota2.com/research/sota/output-prediction-on-cruxeval>)

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

Data updated: 2026-04-21T09:20:54.642Z

Evaluation of model performance on the CruxEval dataset focusing on Output Prediction, measured using Pass@1 metrics differentiated by step return and breakpoint conditions.

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 | Output Prediction |
| Dataset (including any stated split) | CruxEval |
| Metric | Pass@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Pass@1](<https://www.sota2.com/research/sota/output-prediction-on-cruxeval.md?metric=Pass%401>): Higher is better; unit: PERCENT.

- [Pass@1 (step\_return)](<https://www.sota2.com/research/sota/output-prediction-on-cruxeval.md?metric=Pass%401%20(step_return)>): Higher is better; unit: PERCENT.

- [Pass@1 (breakpoint)](<https://www.sota2.com/research/sota/output-prediction-on-cruxeval.md?metric=Pass%401%20(breakpoint)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pass@1 | Result date | Report paper |
| --- | --- | --- | --- |
| DeepSeek-R1 | 94.13 | Not recorded | 69e704bc7d0172140ec7ee7c |
| QwQ-32B | 82.63 | Not recorded | 69e704bc7d0172140ec7ee7c |
| DeepSeek-R1 | 79.25 | Not recorded | 69e704bc7d0172140ec7ee7c |
| DeepSeek-R1 | 79.13 | Not recorded | 69e704bc7d0172140ec7ee7c |
| QwQ-32B | 74.5 | Not recorded | 69e704bc7d0172140ec7ee7c |
| QwQ-32B | 72.38 | Not recorded | 69e704bc7d0172140ec7ee7c |

### DeepSeek-R1

Pass@1: 94.13. Result ID: 69e7416947f34ce924cdc381.

[Reporting paper](<https://www.sota2.com/research/paper/69e704bc7d0172140ec7ee7c.md>)

Recorded evaluation configuration: {"Composition Level":"L1"}. This record does not assert independent reproduction.

### QwQ-32B

Pass@1: 82.63. Result ID: 69e7416947f34ce924cdc384.

[Reporting paper](<https://www.sota2.com/research/paper/69e704bc7d0172140ec7ee7c.md>)

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

Recorded evaluation configuration: {"Composition Level":"L1"}. This record does not assert independent reproduction.

### DeepSeek-R1

Pass@1: 79.25. Result ID: 69e7416947f34ce924cdc383.

[Reporting paper](<https://www.sota2.com/research/paper/69e704bc7d0172140ec7ee7c.md>)

Recorded evaluation configuration: {"Composition Level":"L3"}. This record does not assert independent reproduction.

### DeepSeek-R1

Pass@1: 79.13. Result ID: 69e7416947f34ce924cdc382.

[Reporting paper](<https://www.sota2.com/research/paper/69e704bc7d0172140ec7ee7c.md>)

Recorded evaluation configuration: {"Composition Level":"L2"}. This record does not assert independent reproduction.

### QwQ-32B

Pass@1: 74.5. Result ID: 69e7416947f34ce924cdc385.

[Reporting paper](<https://www.sota2.com/research/paper/69e704bc7d0172140ec7ee7c.md>)

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

Recorded evaluation configuration: {"Composition Level":"L2"}. This record does not assert independent reproduction.

### QwQ-32B

Pass@1: 72.38. Result ID: 69e7416947f34ce924cdc386.

[Reporting paper](<https://www.sota2.com/research/paper/69e704bc7d0172140ec7ee7c.md>)

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

Recorded evaluation configuration: {"Composition Level":"L3"}. This record does not assert independent reproduction.

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