---
title: "Solo Creator Grafts Z-Image Textures Into MiniMax-H3 Without Any Training"
subtitle: "A community-made checkpoint grafts Z-Image's texture attention onto MiniMax H3's video engine, giving richer surfaces without retraining or extra VRAM."
url: "https://alphasignal.ai/news/solo-creator-grafts-z-image-textures-into-minimax-h3-without-any-training"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-08-22T16:13:35.000Z"
updated: "2026-08-27T20:00:20.978Z"
type: "Model"
topics:
  - "Video"
  - "Gpus"
subtopics:
  - "Video Generation"
  - "Quantization"
upvotes: 32
source_url: "https://huggingface.co/joeygambino/MiniMax-H3-x-Z-Image-native"
comments: 0
truncated: true
---

# Solo Creator Grafts Z-Image Textures Into MiniMax-H3 Without Any Training

> A community-made checkpoint grafts Z-Image's texture attention onto MiniMax H3's video engine, giving richer surfaces without retraining or extra VRAM.

A solo creator on Hugging Face has quietly shipped one of the more interesting model surgery experiments this year: a merge that transplants the spatial attention behavior of an image model onto a video model, with no training involved. The result, [MiniMax-H3 x Z-Image-native](https://huggingface.co/joeygambino/MiniMax-H3-x-Z-Image-native), is a drop-in checkpoint for ComfyUI users who want richer textures out of MiniMax H3 without giving up speed, VRAM, or identity fidelity.

## The graft, explained

MiniMax H3 is MiniMax's general-purpose, omni-modal generation model, now available as open weights. It jointly understands text, images, video, and audio in a single context, and generates video with native stereo audio: voice, sound effects, and music are modeled together in a single forward pass instead of layered on afterward. Output tops out around 2K resolution, 24fps, and about 15 seconds. Z-Image, the donor here, is a 6B Lumina2-based image model prized for how sharply it renders fine surface texture.

Both models happen to use per-head Q normalization in attention, and the author, joeygambino, exploits that shared plumbing. Rather than copying weights across, the graft transplants the

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [MiniMax-H3-x-Z-Image-native on Hugging Face](https://huggingface.co/joeygambino/MiniMax-H3-x-Z-Image-native)
- [MiniMax-H3-x-Z-Image-GGUF on Hugging Face](https://huggingface.co/joeygambino/MiniMax-H3-x-Z-Image-GGUF)
- [MiniMax H3 in ComfyUI: T2V, I2V, and R2V Video Workflows](https://docs.comfy.org/tutorials/video/minimax/minimax-h3)
- [MiniMax H3 Day-0 Support in ComfyUI](https://blog.comfy.org/p/minimax-h3-day-0-support-in-comfyui)