Downloads · 30 days
16.6K
29% of all-time downloads
MiniMax-H3-x-Z-Image-native
joeygambino/MiniMax-H3-x-Z-Image-native
MiniMax-H3-x-Z-Image-native is a text-to-video model from joeygambino. Use it when you need video from a text prompt. It is set up for minimax-h3. The card lists the license as other.
The comfy-native cuts of the MiniMax-H3 × Z-Image graft: Z-Image's spatial-attention profile on H3's engine — richer sets and textures, same identity, no per-shot sharpening creep. Full story, demos and verification o…
All-time downloads
57.2K
Public
Repo size
305 GB
Likes
105
Trending 1
Repo makeup
Click a slice to open those files.
.safetensors305 GB · 100%
At a glance
- Task
- Text-to-Video
- Library
- minimax-h3
- License
- other
- Access
- Public
- Created
- Aug 22, 2026
- Updated
- Aug 24, 2026
- SHA
- 113de7b8
Base models
Model card
From the Hugging Face model README
MiniMax-H3 × Z-Image — spatial detail graft (comfy-native)
The comfy-native cuts of the MiniMax-H3 × Z-Image graft: Z-Image's spatial-attention profile on H3's engine — richer sets and textures, same identity, no per-shot sharpening creep. Full story, demos and verification on the GGUF page.
Load with the plain Load Diffusion Model node, ComfyUI 0.32+. Files are the pruned
H3 builds with the graft baked in (zs05 = late-block gains, dose 0.5):
bf16— the master (ref2va)comfy-fp8/fp8e5m2— fp8 scaledcomfy-int8/int8_convrot— the fast pick on RTX 50comfy-w4a8/w4a4/nvfp4— 4-bit family for 16 GB cards (w4a8is the quality pick;nvfp4is Blackwell-native, emulated elsewhere)comfy-mxfp8— 8-bit microscaling, Blackwell-specialized; little benefit on older architectures- fl2va and ref2va variants where both exist; picking rules are identical to stock H3
RTX 30/40: the GGUF repo is 4–8× faster than any 4-bit comfy-native arm on Ampere.
Made by joeygambino. Questions: open a discussion — I answer.
- Task
- Text-to-Video
- Library
- minimax-h3
- License
- other
- Created
- Aug 22, 2026
- Updated
- Aug 24, 2026