Benchmarks · creative
AA-3D
AA-3D: agents in an offline sandbox turn a written brief into an interactive three.js scene; Elo from blind pairwise votes by AA's recruited panel.
21 models · 21 results · data as of 2026-11-06 · artificialanalysis.ai ↗
Leaderboard
Sonnet 5.5 max
More from this benchmark
12 published measuresScore vs any published measure12 measures: sub-scores, splits, counts
10 more models under All models in settings.
Breakdown
darker is better within each column · click a column to sort| Model | |||||
|---|---|---|---|---|---|
| Sonnet 5.5 max | 108.6 | 139.0 | 160.0 | 192.0 | 237.8 |
| Opus 5.5 max | 76.4 | 109.5 | 137.0 | 158.0 | 207.0 |
| GPT-6 Astra max | 42.0 | 51.0 | 60.0 | 69.5 | 76.0 |
| Gemini 4 Argon high | 31.7 | 39.0 | 46.0 | 58.0 | 80.5 |
| Fable 5.1 max | 90.3 | 115.0 | 138.5 | 157.5 | 195.4 |
| GPT-6 Sol max | 53.1 | 65.5 | 73.0 | 89.0 | 111.3 |
| GPT-6.1 Sol max | 39.7 | 48.0 | 56.0 | 63.0 | 68.9 |
| Kimi K3.1 max | 70.1 | 90.0 | 100.0 | 112.0 | 131.2 |
| Muse Spark 1.3 max | 52.5 | 69.5 | 82.0 | 100.0 | 123.0 |
| Inkling 2 Small | 89.7 | 126.5 | 175.0 | 198.0 | 250.0 |
| GPT-6 Luna max | 46.1 | 57.5 | 67.0 | 86.0 | 116.0 |
| Haiku 5.5 max | 39.0 | 47.5 | 57.0 | 72.8 | 84.7 |
| Grok 4.7 xhigh | 49.2 | 64.0 | 81.0 | 96.0 | 148.4 |
| MiMo V2.6 Pro | 77.8 | 103.5 | 117.0 | 128.5 | 152.1 |
| Kimi K3 max | 76.4 | 121.0 | 137.0 | 162.5 | 189.8 |
| Gemini 3.8 Flash high | 72.0 | 90.0 | 109.0 | 122.0 | 160.7 |
| GLM 5.3 Flash max | 97.0 | 152.0 | 182.0 | 249.0 | 250.0 |
| GLM 5.3 max | 165.3 | 242.0 | 250.0 | 250.0 | 250.0 |
| MiniMax M3 | 126.2 | 160.5 | 228.0 | 249.0 | 250.0 |
| Inkling xhigh | 42.7 | 52.5 | 65.0 | 86.0 | 109.9 |
| Nemotron 3 Ultra max | 31.0 | 48.5 | 79.5 | 164.3 | 250.0 |
All results
21 rows| # | Model | Effort | AA-3D | Harness |
|---|---|---|---|---|
| 1 | Sonnet 5.5 | max | 1199 ±33 | Stirrup |
| 2 | Opus 5.5 | max | 1196 ±33 | Stirrup |
| 3 | GPT-6 Astra | max | 1164 ±32 | Stirrup |
| 4 | Gemini 4 Argon | high | 1153 ±33 | Stirrup |
| 5 | Fable 5.1 | max | 1135 ±32 | Stirrup |
| 6 | GPT-6 Sol | max | 1121 ±32 | Stirrup |
| 7 | GPT-6.1 Sol | max | 1119 ±32 | Stirrup |
| 8 | Kimi K3.1 | max | 1110 ±33 | Stirrup |
| 9 | Muse Spark 1.3 | max | 1033 ±31 | Stirrup |
| 10 | Inkling 2 Small | 1021 ±34 | Stirrup | |
| 11 | GPT-6 Luna | max | 1000 ±0 | Stirrup |
| 12 | Haiku 5.5 | max | 998 ±31 | Stirrup |
| 13 | Grok 4.7 | xhigh | 995 ±31 | Stirrup |
| 14 | MiMo V2.6 Pro | 983 ±31 | Stirrup | |
| 15 | Kimi K3 | max | 982 ±31 | Stirrup |
| 16 | Gemini 3.8 Flash | high | 966 ±31 | Stirrup |
| 17 | GLM 5.3 Flash | max | 937 ±31 | Stirrup |
| 18 | GLM 5.3 | max | 783 ±34 | Stirrup |
| 19 | MiniMax M3 | 740 ±33 | Stirrup | |
| 20 | Inkling | xhigh | 583 ±40 | Stirrup |
| 21 | Nemotron 3 Ultra | max | 573 ±43 | Stirrup |