Baroque oil painting test: GPT Image best brushwork, hands still tangled
aziz4ai · x · 2026-11-11
Round 2 of aziz4ai's image-model comparison: a baroque oil painting prompt specifying six people and a single candle. GPT Image 2.6 had the best brushwork and shadow, but some hands still came out overlapping — hand anatomy remains a stubborn failure mode even on the latest models.
More from Multimodal
- MiniMax H3 Image day-one tips: one checkpoint for t2i and edits — ggbro_itsover · 2026-11-11
- MiniMax H3-Image tested: 8.41MP images in ~35s on a 16GB GPU — mininuvixi · 2026-11-11
- Martine benchmark debuts for multi-view 3D surface reconstruction across viewpoints and lighting — ssh4net · 2026-11-11
- Dev builds explorable 3D worlds where a paper koi travels through three lands — techartist_ · 2026-11-11
- H3-FaceRefine-Plus doubles face detail on small faces, open-sourced with full comparisons — AnniCodex · 2026-11-11
- One sentence to an 8-second video: Seedance 2 Mini turns a cat knight into motion — LinTorch · 2026-11-11