Voxel pagoda benchmark pits GPT-6.2 against Opus 5.5, Gemini 4, Kimi K3.1
cedric_chee · x · 2026-11-18
The author extended his 3D voxel pagoda test into a cross-vendor comparison: GPT-6.2 Astra/Sol vs Opus 5.5, Sonnet 5.5, Gemini 4 Hydrogen, DeepSeek-V4.1-Flash and Kimi K3.1 Pro. Screenshots show strikingly different styles at high/xhigh effort — GPT-6.2 Astra and Gemini 4 Hydrogen excel at lighting and night scenes, Opus 5.5 is color-rich, DeepSeek-V4.1-Flash leans ukiyo-e aesthetics, and Kimi K3.1 Pro holds up well at high/max. A rare same-task visual quality comparison across six frontier models.
More from Models
- AI agents sometimes produce unreadable but correct math — burny_tech · 2026-11-18
- Code model eval: Rust hits 95-100% accuracy as 13 languages see big gains — shuding · 2026-11-18
- Asked Opus to make a video explaining consciousness and existence — "no wiser but cool" — flowersslop · 2026-11-18
- theo retests GPT-6.2 Astra: better, but still 'directionally very bad' at debugging — airesearch12 · 2026-11-18
- Claude correctly folds ubiquitin from a plain prompt, no specialized tools needed — iScienceLuvr · 2026-11-18
- GPT-6.2 Astra Tops Bug-Hunt Coding Benchmark at 62.7/105, One-Fifth Sonnet's Cost — PawelHuryn · 2026-11-18