Design Arena (ASCII Art): leaderboard
Design Arena sends one prompt to four anonymous models and asks visitors to pick the better output in a series of pairwise comparisons. This board is the ASCII art category, where the artefact is text drawn with characters. Votes are fitted with a Bradley Terry model and reported as an Elo style rating, and a model needs at least fifteen comparisons before it is listed.
Metric: Elo. Source: www.designarena.ai. Saturation forecast: Around December 2027. 95 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 1356 | #6 |
| 2 | Claude Opus 5 | 1352 | #12 |
| 3 | Claude Fable 5 | 1329 | #16 |
| 4 | GPT-6 Astra (Max) | 1329 | #6 (GPT-6 Astra) |
| 5 | Claude Fable 5.1 | 1323 | #9 |
| 6 | GPT-5.6 Sol (xHigh) | 1300 | #19 (GPT-5.6 Sol) |
| 7 | Muse Spark 1.2 | 1285 | #62 |
| 8 | Muse Spark 1.3 | 1283 | #24 |
| 9 | Gemini 3.6 Flash | 1279 | #51 |
| 10 | Claude Opus 4.8 | 1277 | #37 |
| 11 | Gemini 3.7 Flash | 1276 | #23 |
| 12 | Gemini 3.1 Pro (Preview) | 1270 | #58 |
| 13 | DeepSeek V4.1 Flash | 1265 | #36 |
| 14 | Gemini 3.8 Flash | 1262 | #21 |
| 15 | GPT-5.5 | 1256 | #31 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=design-arena-ascii-art · How It Works · Data refreshed daily, snapshot 2026-10-20.