Benchmark Prompts / P02
Desk lamp: 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.
12Outputs643 ballots · 8 both-bad
A desk lamp with an articulated two-joint arm and a round metal shade, in a bent working pose.
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 5.5Desk lamp Grok 4.7Desk lamp GPT-6 SolDesk lamp GPT-6 LunaDesk lamp GPT-6.1 SolDesk lamp 3D-AgentDesk lamp GPT-6 AstraDesk lamp Claude Fable 5.1Desk lamp Claude Opus 5Desk lamp GPT-5.6 SolDesk lamp Grok 4.6Desk lamp Claude Fable 5Desk lamp
Season 3 blind votes for P02
643 ballots · 8 “both bad”
| Competitor | Record | Wins | Losses |
|---|---|---|---|
| Claude Opus 5.5 | 72 | 24 | |
| GPT-6 Astra | 63 | 38 | |
| 3D-Agent | 62 | 42 | |
| GPT-6 Sol | 60 | 31 | |
| Grok 4.6 | 59 | 48 | |
| Grok 4.7 | 58 | 43 | |
| Claude Opus 5 | 54 | 55 | |
| GPT-6.1 Sol | 53 | 7 | |
| Claude Fable 5.1 | 47 | 50 | |
| GPT-5.6 Sol | 42 | 63 | |
| GPT-6 Luna | 16 | 71 | |
| Claude Fable 5 | 2 | 116 |