blazing fast, only 1 minute 10 seconds generating 5 second clip in 0.4 mp resolution.
NEW using Ultimate upscale 1.0mp resolution or 768p got 3 minutes in my rtx 4060ti 16gb
UPDATE!!!:
i ditch sage attention and use Comfy kitchen + sol attention instead, but i'm still use kjnodes. here is the link to install custom nodes:
kjnodes: https://github.com/kijai/ComfyUI-KJNodes
Comfy Kitchen: https://github.com/Comfy-Org/comfy-kitchen
Sol attention Triton: https://github.com/kijai/ComfyUI-SolAttn_triton
use this model from huggingface: https://huggingface.co/MATLOWAI/minimax-h3-fused-turbo-int8-convrot/tree/main/diffusion_models
if you like my work and want to support me, you can donate me to my ko-fi: https://ko-fi.com/megaaziib
get minimax ultimate upscale: https://github.com/bbaudio-2025/Comfyui-MMH3-UltimateUpscale
for upscale model, you need to set the second resolution either 0.6, 0.7- 1.0. megapixel. if you get oom or generating too long, try to decrease resolution to 0.9 or 0.8 mp.
change manualsigmas according your preference steps:
4 steps:
0.9999166, 0.9728326, 0.9230769, 0.8, 0.0
6 steps:
0.9999166, 0.9868421, 0.9638554, 0.9230769, 0.8695652, 0.8, 0.0
8 steps:
1.0, 0.9882352941, 0.9729729730, 0.9523809524, 0.9230769231, 0.8780487805, 0.8, 0.6315789474, 0.0
Description
FAQ
Comments (7)
4060Ti 16G 32Ram,0.5 ITV 5 seconds 407S
what clip are u loading? for that 4060ti
qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
should be available on comfyui huggingface: https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/text_encoders
Excellent workflow. Everyone should give it a try
1min 33sec on a 5070 ti 16gb, no sage attention enabled, 3 loras loaded.
When I first tested this I had issues with the KJ Diffusion Model loader having a fault. So, I replaced it with the KJ Diffusion model loader. It worked. Weird. I ran the workflow. It ran fast. It ran well. And it seems to follow prompts better than other workflows I've tried.
Now, I might be hallucinating a bit, but it doesn't seem to be as crip at making video as other flows. But it does render t2v way faster than my other workflow. Most importantly, to me, it really follows directions. Yeah, I had to prompt and prompt again to get a few things right. But that's AI.
I am really liking this workflow. One of the better one's I have had the pleasure to play with.
try recent one, i try to change sage attention to comfy kitchen + sol attention.