Community fine-tune fixes Qwen-Image 2.1 VAE 8-pixel grid artifact
Crazy-Repeat-2006 · reddit · 2026-10-24
A Reddit user released a fine-tune of Texture-Fix targeting the 8-pixel grid artifact Qwen-Image 2.1's VAE leaves on fur and fine detail.
- Trains only upblocks.2/.3
- Cuts the 8px frequency peak while raising sharpness
- Later conv layers keep the pattern out of RGB
Weights are on Hugging Face (None1082/None-qwenimage2.1-vae) with a ComfyUI-compatible version by e-n-v-y.
More from Multimodal
- Stunning AI-Generated Sci-Fi Scene: Futuristic Car Rides Toward a Floating Tower — ciguleva · 2026-10-24
- Character LoRA trained on 21 images fails across models—training setup questioned — IntrovertedOtaku03 · 2026-10-24
- Tencent's Hy-Nomas video world model ported to native ComfyUI, runs on 12GB VRAM — LatentSpacer · 2026-10-24
- Video models replace NeRF/3DGS pipelines: single-frame novel view synthesis, open-source soon — jonstephens85 · 2026-10-24
- Recreating PS2-Era Harry Potter With FastH3-V2 in 4 Steps — aziib · 2026-10-24
- A Full AI Music Video Made Locally on DGX Spark With MiniMax H3 — Ornery-Block-3522 · 2026-10-24