MOSS-Image comes to Forge Neo: readable English/Chinese text, 8-step generation
digitalhunters0 · reddit · 2026-10-18
A developer released an open-source extension bringing MOSS-Image to Forge Neo, working in the familiar txt2img/img2img tabs. Key features: text-to-image with readable English and Chinese text (posters, signs, labels) with strong prompt adherence; img2img editing that preserves composition; layout control; support for both SFT and Base models. A highlight is the lodestones 8-steps-dmd LoRA, cutting generation from 50 steps to 8 while keeping good quality and readable text, though fine details still fare better at 50 steps. Installable from Forge Neo's Extensions tab; early release with instructions and examples in the repo.
More from Multimodal
- Why top creators still pick Midjourney: taste is the only moat in AI image generation — EXM7777 · 2026-10-18
- Magnific One image model debuts with eye-catching heel-art generations — charis_ai · 2026-10-18
- Flux3 vs Flux2 9B style transfer: better composition fidelity at 4-5x the speed — Winter_unmuted · 2026-10-18
- Agent spawns its own 'procedural audio composer' subagent and synthesizes a vaporwave track in Python — teortaxesTex · 2026-10-18
- A 3D journey tracing one message from a café table to an overseas data centre — techartist_ · 2026-10-18
- Running Fish Audio Drama 3 on a 16GB AMD GPU: Red Curtain ports the 35B MoE that officially wants 48GB — fechyyy · 2026-10-18