Benchmarks · creative
ClutterBench, simple prompt
Clutter as a share of drawn area in generated graphics, using the simple prompt. Red area plus half the orange area, averaged over 6 scored posts. Lower is better.
Category
11 models · 40 results · updated 2026-10-19 · clutterbench.adamholter.chatgpt.site ↗
Leaderboard
Haiku 5.5 high
All results
40 rows| # | Model | Effort | Score |
|---|---|---|---|
| 1 | Haiku 5.5 | high | 0.0% |
| 2 | Haiku 5.5 | low | 0.0% |
| 3 | Haiku 5.5 | medium | 0.0% |
| 4 | Haiku 5.5 | xhigh | 0.0% |
| 5 | Opus 5.5 | medium | 0.0% |
| 6 | Sonnet 5.5 | high | 0.0% |
| 7 | Sonnet 5.5 | low | 0.0% |
| 8 | DeepSeek V4.1 Flash | high | 0.0% |
| 9 | DeepSeek V4.1 Flash | max | 0.0% |
| 10 | DeepSeek V4.1 Flash | medium | 0.0% |
| 11 | DeepSeek V4.1 Flash | xhigh | 0.0% |
| 12 | Gemini 3.8 Flash | high | 0.0% |
| 13 | Gemini 3.8 Flash | medium | 0.0% |
| 14 | GPT-6 Astra | high | 0.0% |
| 15 | GPT-6 Astra | xhigh | 0.0% |
| 16 | GPT-6 Luna | high | 0.0% |
| 17 | GPT-6 Luna | medium | 0.0% |
| 18 | GPT-6 Luna | xhigh | 0.0% |
| 19 | GPT-6.1 Sol | high | 0.0% |
| 20 | GPT-6.1 Sol | medium | 0.0% |
| 21 | GPT-6.1 Sol | xhigh | 0.0% |
| 22 | Grok 4.7 | high | 0.0% |
| 23 | Grok 4.7 | low | 0.0% |
| 24 | Grok 4.7 | xhigh | 0.0% |
| 25 | DeepSeek V4.1 Flash | low | 0.1% |
| 26 | Gemini 3.8 Flash | low | 0.1% |
| 27 | GPT-6.1 Sol | low | 0.1% |
| 28 | GPT-6 Astra | low | 0.2% |
| 29 | Sonnet 5.5 | xhigh | 0.3% |
| 30 | GPT-6 Astra | medium | 0.9% |