# ASCII Arena

> Kaggle's ASCII Arena asks models to draw everyday subjects as ASCII art, and people vote for the better drawing.

Canonical page: https://benchlm.ai/benchmarks/ascii-arena

- Category: [external](/external)
- Last updated: Latest evaluation November 13, 2026

Evaluation results published by [Kaggle](https://www.kaggle.com/benchmarks/kaggle/ascii-arena). undefined

Subjects come from Google's Quick, Draw! dataset. It tests visual layout using plain text characters only. Kaggle shows two anonymous outputs for the same prompt and a person picks the better one. The Arena column is the relative rating Kaggle fits to those votes, published with an interval; it is not a percentage.

The rating moves as votes arrive, and Kaggle publishes no confidence level for its interval. Rows with few votes have wide intervals, so neighbouring models are often not separable. Voters are self-selected Kaggle visitors. The board declares no license; we keep published aggregates with attribution and no task or output contents.

A rating gap smaller than the two rows' intervals is not a reliable difference.

- [Kaggle arena leaderboard](https://www.kaggle.com/benchmarks/kaggle/ascii-arena)
- [Published leaderboard data](https://www.kaggle.com/api/v1/benchmarks/kaggle/ascii-arena/leaderboard)

## About ASCII Arena

- Year: 2026
- Tasks: ASCII drawings of subjects from Google's Quick, Draw! dataset, rated by human votes
- Format: Relative rating fitted to human pairwise votes
- Difficulty: Text-character drawing

Subjects come from Google's Quick, Draw! dataset. It tests visual layout using plain text characters only. Kaggle shows two anonymous outputs for the same prompt and a person picks the better one. The Arena column is the relative rating Kaggle fits to those votes, published with an interval; it is not a percentage. These results are display-only and do not enter overall or category rankings.

ASCII Arena is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (18 source rows)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Gemini 4 Carbon](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 173 votes; ±127; Evaluated 2026-11-13 | — | 2019 |
| 2 | [Gemini 4 Argon](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 411 votes; ±99; Evaluated 2026-11-11 | — | 1983 |
| 3 | [GPT-6 Astra](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 427 votes; ±86; Evaluated 2026-11-13 | — | 1730 |
| 4 | [GPT-6.1 Sol](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 437 votes; ±85; Evaluated 2026-11-13 | — | 1686 |
| 5 | [Gemini 3.8 Flash](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 405 votes; ±85; Evaluated 2026-11-11 | — | 1672 |
| 6 | [Claude Opus 5.5](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 445 votes; ±84; Evaluated 2026-11-13 | — | 1651 |
| 7 | [Gemini 3.1 Pro Preview](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 454 votes; ±83; Evaluated 2026-11-13 | — | 1573 |
| 8 | [Claude Sonnet 5.5](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 419 votes; ±83; Evaluated 2026-11-13 | — | 1541 |
| 9 | [Claude Opus 5](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 427 votes; ±83; Evaluated 2026-11-12 | — | 1506 |
| 10 | [GPT-6 Sol](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 445 votes; ±82; Evaluated 2026-11-11 | — | 1497 |
| 11 | [Claude Fable 5.1](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 424 votes; ±83; Evaluated 2026-11-13 | — | 1484 |
| 12 | [Claude Haiku 5.5](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 409 votes; ±84; Evaluated 2026-11-12 | — | 1354 |
| 13 | [Kimi K3](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 410 votes; ±84; Evaluated 2026-11-13 | — | 1303 |
| 14 | [GPT-5.6 Terra](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 445 votes; ±84; Evaluated 2026-11-12 | — | 1294 |
| 15 | [Qwen 3.8 Max](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | Xhigh Reasoning; 429 votes; ±84; Evaluated 2026-11-11 | — | 1282 |
| 16 | [GPT-6 Luna](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 447 votes; ±83; Evaluated 2026-11-13 | — | 1279 |
| 17 | [Gemini 3.5 Flash-Lite](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 423 votes; ±84; Evaluated 2026-11-13 | — | 1271 |
| 18 | [GPT-5.6 Luna](https://www.kaggle.com/benchmarks/kaggle/ascii-arena) | High Reasoning; 446 votes; ±84; Evaluated 2026-11-13 | — | 1262 |

## FAQ

### What does ASCII Arena measure?

Kaggle's ASCII Arena asks models to draw everyday subjects as ASCII art, and people vote for the better drawing.

### Which source row has the highest reported ASCII Arena value?

Gemini 4 Carbon has the highest reported ASCII Arena value at 2019. A rating gap smaller than the two rows' intervals is not a reliable difference.

### How many models are evaluated on ASCII Arena?

The Latest evaluation November 13, 2026 contains 18 AI models.

### Does ASCII Arena affect BenchLM's overall score?

Not directly. ASCII Arena is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.