Two-Step LoRA for H3 Text-to-Image Released, Nears 8-Step Baseline Quality

A new 2-step pDMD LoRA for H3 text-to-image generation has been released on Hugging Face, requiring no CFG and producing quality in 4 steps that approaches the 8-step baseline, dramatically speeding up inference.

2026-11-07 ~ 2026-11-07 · 2 related posts

• 2-step pDMD LoRA for p-h3-image released, fix blurry fast-generation outputs — Illustrious_Row_9971 · 2026-11-07
• 2-step LoRA for H3 image model hits 8MP without CFG, near baseline at 4 steps — _akhaliq · 2026-11-07