Benchmark Prompts / P11
Spiral staircase: 12 AI models, one prompt
Every Season 3 competitor received this prompt verbatim, at default settings, with one generation and no manual cleanup. Nothing below is cherry-picked — the failures ship too, because a benchmark that hides them isn't one.
12Outputs615 ballots · 3 both-bad
A spiral staircase with twelve steps winding around a central column, with a handrail.
Every output for this prompt
Click a tile to load the model in 3D, then switch to Wire or X-ray to read the topology. Vote on these blind →
Claude Opus 5.5Spiral staircase Grok 4.7Spiral staircase GPT-6 SolSpiral staircase GPT-6 LunaSpiral staircase GPT-6.1 SolSpiral staircase 3D-AgentSpiral staircase GPT-6 AstraSpiral staircase Claude Fable 5.1Spiral staircase Claude Opus 5Spiral staircase GPT-5.6 SolSpiral staircase Grok 4.6Spiral staircase Claude Fable 5Spiral staircase
Season 3 blind votes for P11
615 ballots · 3 “both bad”
| Competitor | Record | Wins | Losses |
|---|---|---|---|
| GPT-6 Astra | 75 | 26 | |
| 3D-Agent | 72 | 38 | |
| GPT-6 Sol | 64 | 29 | |
| Claude Opus 5.5 | 63 | 21 | |
| Grok 4.7 | 59 | 45 | |
| GPT-6.1 Sol | 58 | 13 | |
| GPT-5.6 Sol | 57 | 44 | |
| Claude Fable 5.1 | 38 | 67 | |
| Claude Opus 5 | 29 | 59 | |
| Grok 4.6 | 23 | 60 | |
| GPT-6 Luna | 18 | 73 | |
| Claude Fable 5 | 9 | 90 |