# Step Prediction on Pen-Strategist (test)

Source: [Step Prediction on Pen-Strategist (test)](<https://www.sota2.com/research/sota/step-prediction-on-pen-strategist-test>)

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

Data updated: 2026-05-07T08:36:16.780Z

Evaluation of model performance on the Step Prediction task using the test split of the Pen-Strategist dataset, measured by Accuracy and Micro F1.

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 | Step Prediction |
| Dataset (including any stated split) | Pen-Strategist (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/step-prediction-on-pen-strategist-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Micro F1 Score](<https://www.sota2.com/research/sota/step-prediction-on-pen-strategist-test.md?metric=Micro%20F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Step Model | 82.87 | Not recorded | 69fc1cac5497b3b670c88619 |
| Gemini 2.5 Flash | 64.52 | Not recorded | 69fc1cac5497b3b670c88619 |
| Claude 4.5 sonnet | 63.44 | Not recorded | 69fc1cac5497b3b670c88619 |
| GPT 4.1 | 62.78 | Not recorded | 69fc1cac5497b3b670c88619 |
| Gemini 2.0 Flash | 61.29 | Not recorded | 69fc1cac5497b3b670c88619 |
| GPT-5-mini | 59.66 | Not recorded | 69fc1cac5497b3b670c88619 |
| Claude 3 Haiku | 55.43 | Not recorded | 69fc1cac5497b3b670c88619 |
| GPT-4o-mini | 54.83 | Not recorded | 69fc1cac5497b3b670c88619 |
| LLaMA-3.1-8B | 51.09 | Not recorded | 69fc1cac5497b3b670c88619 |
| GPT-3.5-turbo | 41.76 | Not recorded | 69fc1cac5497b3b670c88619 |

### Step Model

Accuracy: 82.87. Result ID: 69fc4dae47f34ce924df026c.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Gemini 2.5 Flash

Accuracy: 64.52. Result ID: 69fc4dae47f34ce924df0266.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78731e209131cea78622.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Claude 4.5 sonnet

Accuracy: 63.44. Result ID: 69fc4dae47f34ce924df0264.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPT 4.1

Accuracy: 62.78. Result ID: 69fc4dae47f34ce924df0268.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Gemini 2.0 Flash

Accuracy: 61.29. Result ID: 69fc4dae47f34ce924df0265.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78731e209131cea78622.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPT-5-mini

Accuracy: 59.66. Result ID: 69fc4dae47f34ce924df026a.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Claude 3 Haiku

Accuracy: 55.43. Result ID: 69fc4dae47f34ce924df0263.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPT-4o-mini

Accuracy: 54.83. Result ID: 69fc4dae47f34ce924df0269.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LLaMA-3.1-8B

Accuracy: 51.09. Result ID: 69fc4dae47f34ce924df026b.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPT-3.5-turbo

Accuracy: 41.76. Result ID: 69fc4dae47f34ce924df0267.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cac5497b3b670c88619.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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