MOSS-Image ecosystem roundup: smart sampler, handwriting LoRAs, 7.5GB WebGPU build
optimisticalish · reddit · 2026-11-03
A community roundup of new MOSS-Image tooling: a ComfyUI Smart Sampler with Stage Weighted/Smooth Warp sigma curves and presets; a LoRA trained on 17th–19th century Nordic archival scans rendering eight historic handwriting styles; a manga panel-layout prompt builder node; an un-GPT LoRA for flatter shading and cleaner linework in anime and icons; and MossInt4, a 4-step SFT squeezed to 7.5GB running fully in-browser via WebGPU.
More from Multimodal
- AI Video Fails Double Feature: Sleepwalking Kid and the Runaway Coffee Machine — nastyalarm · 2026-11-03
- 12-step Krea 2-distilled LoRA pushes Anima to photorealism, with full pipeline and fixes — Honest_Concert_6473 · 2026-11-03
- Runway launches Runway Dev, an API platform built for coding agents — runwayml · 2026-11-03
- Single-shot 2-minute drive video with MiniMax H3/Slopus, start frame as only reference — BittiAI · 2026-11-03
- Creator shares full timestamped prompt for photoreal southern gothic swamp creature AI video — LudovicCreator · 2026-11-03
- Industrial designer uses fal to iterate real hardware designs from photos, faster prototyping — OdinLovis · 2026-11-03