# Creative Writing Evaluation on EQ-Bench Creative (OOD)

Source: [Creative Writing Evaluation on EQ-Bench Creative (OOD)](<https://www.sota2.com/research/sota/creative-writing-evaluation-on-eq-bench-creative-ood>)

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

Data updated: 2026-06-04T10:56:06.366Z

Evaluation of model performance on the Out-of-Distribution (OOD) subset of the EQ-Bench Creative dataset for creative writing tasks, measured using the Elo rating system.

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 | Creative Writing Evaluation |
| Dataset (including any stated split) | EQ-Bench Creative (OOD) |
| Metric | Elo Rating |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [Elo Rating](<https://www.sota2.com/research/sota/creative-writing-evaluation-on-eq-bench-creative-ood.md?metric=Elo%20Rating>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Elo Rating | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-5.4 | 1911 | Not recorded | 6a2106dec10484bac23f0524 |
| Claude Opus 4.6 | 1783 | Not recorded | 6a2106dec10484bac23f0524 |
| POLARIS–9B | 1661 | Not recorded | 6a2106dec10484bac23f0524 |
| Gemini 3.1 Pro | 1627 | Not recorded | 6a2106dec10484bac23f0524 |
| Gemini 3 Flash | 1620 | Not recorded | 6a2106dec10484bac23f0524 |
| Gemma-4-31B | 1514 | Not recorded | 6a2106dec10484bac23f0524 |
| Qwen3.5-27B | 1503 | Not recorded | 6a2106dec10484bac23f0524 |
| Gemma-4-26B A4B\* | 1460 | Not recorded | 6a2106dec10484bac23f0524 |
| Qwen3.5-9B | 1352 | Not recorded | 6a2106dec10484bac23f0524 |
| Gemma-4-E4B† | 1312 | Not recorded | 6a2106dec10484bac23f0524 |
| Ministral-3-8B | 1219 | Not recorded | 6a2106dec10484bac23f0524 |
| DeepSeek-R1-Distill-Qwen-14B | 1030 | Not recorded | 6a2106dec10484bac23f0524 |
| LongWriter-Zero-32B | 1013 | Not recorded | 6a2106dec10484bac23f0524 |
| LongWriter-Llama3.1-8B | 800 | Not recorded | 6a2106dec10484bac23f0524 |

### GPT-5.4

Elo Rating: 1911. Result ID: 6a2159b6fbc373592c1de32e.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Frontier","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Claude Opus 4.6

Elo Rating: 1783. Result ID: 6a2159b6fbc373592c1de32f.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Frontier","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### POLARIS–9B

Elo Rating: 1661. Result ID: 6a2159b6fbc373592c1de330.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Gemini 3.1 Pro

Elo Rating: 1627. Result ID: 6a2159b6fbc373592c1de331.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Frontier","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Gemini 3 Flash

Elo Rating: 1620. Result ID: 6a2159b6fbc373592c1de332.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

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

Recorded evaluation configuration: {"Model Category":"Frontier","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Gemma-4-31B

Elo Rating: 1514. Result ID: 6a2159b6fbc373592c1de333.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Qwen3.5-27B

Elo Rating: 1503. Result ID: 6a2159b6fbc373592c1de334.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Gemma-4-26B A4B\*

Elo Rating: 1460. Result ID: 6a2159b6fbc373592c1de335.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Architecture":"MoE","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Qwen3.5-9B

Elo Rating: 1352. Result ID: 6a2159b6fbc373592c1de336.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Gemma-4-E4B†

Elo Rating: 1312. Result ID: 6a2159b6fbc373592c1de337.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Architecture":"PLE","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### Ministral-3-8B

Elo Rating: 1219. Result ID: 6a2159b6fbc373592c1de338.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### DeepSeek-R1-Distill-Qwen-14B

Elo Rating: 1030. Result ID: 6a2159b6fbc373592c1de339.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### LongWriter-Zero-32B

Elo Rating: 1013. Result ID: 6a2159b6fbc373592c1de33a.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a31e209131cea7e984.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

### LongWriter-Llama3.1-8B

Elo Rating: 800. Result ID: 6a2159b6fbc373592c1de33b.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0524.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa746e1e209131cea61ae6.md>)

Recorded evaluation configuration: {"Model Category":"Open-weight","Judge":"GPT-5.4"}. This record does not assert independent reproduction.

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