Benchmark Prompts / P03
Low-poly knight: 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.
12Outputs604 ballots · 30 both-bad
A low-poly stylized knight character in idle pose, with helmet, shield on back, T-pose acceptable.
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 5.5Low-poly knight Grok 4.7Low-poly knight GPT-6 SolLow-poly knight GPT-6 LunaLow-poly knight GPT-6.1 SolLow-poly knight 3D-AgentLow-poly knight GPT-6 AstraLow-poly knight Claude Fable 5.1Low-poly knight Claude Opus 5Low-poly knight GPT-5.6 SolLow-poly knight Grok 4.6Low-poly knight Claude Fable 5Low-poly knight
Season 3 blind votes for P03
604 ballots · 30 “both bad”
| Competitor | Record | Wins | Losses |
|---|---|---|---|
| Claude Opus 5.5 | 78 | 8 | |
| GPT-6.1 Sol | 75 | 3 | |
| Claude Opus 5 | 65 | 49 | |
| GPT-6 Sol | 61 | 15 | |
| GPT-5.6 Sol | 57 | 51 | |
| 3D-Agent | 55 | 34 | |
| Claude Fable 5.1 | 41 | 33 | |
| GPT-6 Luna | 41 | 39 | |
| GPT-6 Astra | 33 | 64 | |
| Grok 4.6 | 33 | 51 | |
| Claude Fable 5 | 17 | 100 | |
| Grok 4.7 | 2 | 111 |