MOSS-Image Round-up: GrabTag Region Control, New LoRAs and a Layout Cookbook
optimisticalish · reddit · 2026-11-07
A MOSS-Image community round-up for November 7:
- GrabTag: turns a reference character image into region-controlled input — auto-segments parts, tags each, and outputs a region-ID color map plus layout JSON; ComfyUI guidance included
- An 8GB "amateur look" slider LoRA (no trigger, strength 1.5) that also roughens skin
- Cinema Wide Lens: widescreen letterbox bars with blue night flares
- Pen Sketch Style v1: ballpoint scribble style LoRA
- NeoObscura: modern surreal expressionist illustration LoRA for posters
- A YouTube tutorial on fixing wrong words in posters, and a Layout Cookbook with 12 copy-paste layout JSON examples
More from Multimodal
- Seedance 2.5's pyramid-building video is so realistic it's fooling viewers — CurieuxExplorer · 2026-11-08
- PD Forma: 4D 3D reconstruction from vehicle and drone sensor data — RexDouglass · 2026-11-07
- PixAI V4.0 stress test: 15-second mecha battle clips that survive fast action — Promptmethus · 2026-11-07
- Chat-driven filmmaking with MiniMax H3: eight scene recipes and a continuity test — Eastern-Ad9515 · 2026-11-07
- RawSnap v3 LoKr released: retrained on 749 images, stackable with v2 for realism — Wide_Director_8897 · 2026-11-07
- MiniMax-H3 compressed to run video+audio in 6 steps on 24GB GPUs — Cool-Door3047 · 2026-11-07