Open Image Base Model Moss Hailed as Best Yet, Trained on Uncensored Data
1or4s · reddit · 2026-10-18
A Reddit user praises the new open image base model Moss as the best base model so far: where Krea 2 struggles even with LoRAs and imposes an inflexible style, Moss handles out of the box. Notably, Moss is not only uncensored at inference—its training data itself included uncensored images, which is rare for a base model and suggests very high community finetune potential. "The golden age has arrived," the author writes.
More from Multimodal
- Dev builds Mumbai-inspired 3D manga background generator from OpenStreetMap data — CurieuxExplorer · 2026-10-18
- Omni 1'1 flash video demo generates atmospheric foggy Victorian haunted house — Dense-Instruction210 · 2026-10-18
- Flux 3 klein distilled tested: int8 workflow delivers surprisingly good 2MP images — Ok_Tale7582 · 2026-10-18
- AI-generated clip with cinematic truck-driver shot fuels 'feature film soon' talk — 1ili0 · 2026-10-18
- AI Videos Are Getting So Good That Real Content Now Gets Doubted — gouterz · 2026-10-18
- Qwen-Image-2.1 LoRA Unlocks Nearly Any Pose, Works for Both T2I and Editing — metalfans · 2026-10-18