Instructions to use Comfy-Org/MiniMax-H3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use Comfy-Org/MiniMax-H3 with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
[ Ranking ] H3 Video VAE Speed Comparison ! 🏅
#69
by sunnyboxs - opened
视频时长 = 10秒, 解码分辨率 = 1664x928
Video duration = 10 seconds, decoding resolution = 1664x928
(1) 🥇 minimax_h3_sol_vae_int8_convrot.safetensors
Peak VRAM usage: 9 GB, duration: 13 seconds
显存峰值9GB, 耗时=13秒 (winner)
(2) 🥈minimax_h3_lightvae_lite_vae_int8_convrot.safetensors
[Lower image quality] Peak VRAM usage: 5 GB, duration: 17 seconds
[画质较差]显存峰值5GB, 耗时=17秒
(3) 🥉minimax_h3_lynnreal_light_vae_int8_convrot.safetensors
Peak VRAM usage: 5 GB, duration: 35 seconds
显存峰值5GB, 耗时=35秒
(4) 🎖️minimax_h3_video_vae_int8_convrot.safetensors
Peak VRAM usage: 5.5 GB, duration: 48 seconds
显存峰值5.5GB, 耗时=48秒
请问这个vae节点怎样使用?我替换了原来的minimax_h3_video_vae_int8_convrot.safetensors会报错
请问这个vae节点怎样使用?我替换了原来的minimax_h3_video_vae_int8_convrot.safetensors会报错
更新comfyui到最新版
Is there a difference in image quality?
sol gives me red eroor

