# Arena vision style control / creative writing

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Acreative_writing. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Acreative_writing.

Human pairwise preferences on creative writing prompts in Arena’s vision style control evaluation. The original organizer publishes model ratings and uncertainty, adjusting for response style; these are preference ratings rather than task accuracy.

Canonical URL: https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Acreative_writing
Last updated: Wed, 21 Oct 2026 18:05:10 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Human pairwise preferences on creative writing prompts in Arena’s vision style control evaluation. The original organizer publishes model ratings and uncertainty, adjusting for response style; these are preference ratings rather than task accuracy.

Multimodal · Models tested: 13 · Score publishers: [Arena](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)

## Model results

| # | Model | Score |  |
| --- | --- | --- | --- |
| 1 | [Gemini 3 Pro](https://benchize.com/models/gemini-3-pro.md) | 1,372.9 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1372.897919232414 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1350.3665980521116 rank 1 upper 1395.4292404127166 variance 132.15303295590607 votes 766 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gemini-3-pro Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/329/row` |  |  |  |
| 2 | [Gemini 3 Flash](https://benchize.com/models/gemini-3-flash.md) | 1,289.1 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1289.1029615413795 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1248.916708822333 rank 3 upper 1329.2892142604258 variance 420.3962564688528 votes 217 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gemini-3-flash Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/331/row` |  |  |  |
| 3 | [GPT-5.1](https://benchize.com/models/gpt-5-1.md) | 1,284.6 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1284.6038276520364 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1255.1481814483368 rank 4 upper 1314.059473855736 variance 225.86083406739368 votes 411 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gpt-5.1 Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/332/row` |  |  |  |
| 4 | [Gemini 2.5 Pro](https://benchize.com/models/gemini-2-5-pro.md) | 1,262.39 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1262.387331733268 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1248.49317268865 rank 7 upper 1276.281490777886 variance 50.25373551245191 votes 2418 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gemini-2.5-pro Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/335/row` |  |  |  |
| 5 | [GPT-5 Chat](https://benchize.com/models/gpt-5-chat.md) | 1,231.82 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1231.8171281800587 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1215.94773257538 rank 12 upper 1247.6865237847371 variance 65.55783326405334 votes 1518 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gpt-5-chat Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/340/row` |  |  |  |
| 6 | [Gemini 2.5 Flash](https://benchize.com/models/gemini-2-5-flash.md) | 1,230.89 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1230.887538447331 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1215.862144676827 rank 13 upper 1245.9129322178349 variance 58.76997997284331 votes 1745 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gemini-2.5-flash Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/341/row` |  |  |  |
| 7 | [Qwen3 VL 235B A22B Thinking](https://benchize.com/models/qwen3-vl-235b-a22b-thinking.md) | 1,217.58 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1217.579969266169 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1179.3011489613757 rank 14 upper 1255.8587895709625 variance 381.4353224439617 votes 219 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label qwen3-vl-235b-a22b-thinking Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/342/row` |  |  |  |
| 8 | [Gemini 2.0 Flash 001](https://benchize.com/models/gemini-2-0-flash-001.md) | 1,209.11 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1209.1118505446868 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1175.6565536869714 rank 15 upper 1242.5671474024025 variance 291.36245892013136 votes 292 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gemini-2.0-flash-001 Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/343/row` |  |  |  |
| 9 | [Qwen3 VL 235B A22B Instruct](https://benchize.com/models/qwen3-vl-235b-a22b-instruct.md) | 1,193.5 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1193.4976349347016 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1170.9647839505606 rank 19 upper 1216.0304859188427 variance 132.17097909219925 votes 690 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label qwen3-vl-235b-a22b-instruct Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/347/row` |  |  |  |
| 10 | [Gemma 3 27B](https://benchize.com/models/gemma-3-27b-it.md) | 1,174.6 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1174.6045963457343 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1152.6282918089 rank 25 upper 1196.5809008825686 variance 125.72254022195004 votes 732 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label gemma-3-27b-it Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/353/row` |  |  |  |
| 11 | [Llama 4 Scout](https://benchize.com/models/llama-4-scout.md) | 1,137.93 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1137.931183100854 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1101.1497480538835 rank 31 upper 1174.7126181478243 variance 352.17713563047823 votes 265 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label llama-4-scout-17b-16e-instruct Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/359/row` |  |  |  |
| 12 | [Mistral Small 3.1 24B](https://benchize.com/models/mistral-small-3-1-24b-instruct.md) | 1,132.03 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1132.0262216501865 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1110.2175406062745 rank 32 upper 1153.8349026940984 variance 123.81196599398773 votes 777 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label mistral-small-3.1-24b-instruct-2503 Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/360/row` |  |  |  |
| 13 | [Llama 4 Maverick](https://benchize.com/models/llama-4-maverick.md) | 1,127.8 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1127.8039604923083 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config vision_style_control ### Reported statistics lower 1090.048890623918 rank 33 upper 1165.559030360699 variance 371.06874427186034 votes 250 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label llama-4-maverick-17b-128e-instruct Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/361/row` |  |  |  |

---
Cite as: Arena vision style control / creative writing — Benchize, https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Acreative_writing (retrieved 2026-10-21)