Kijai
/

MiniMax-H3-experimental

Model card Files Files and versions

Resources

View closed (5)

Unpruned W4A8 Diffusion Model Request

#27 opened 4 days ago by

makisekurisu-jp

The smaller models slow my generation down

#26 opened 5 days ago by

Reference Video

#25 opened 5 days ago by

works well - NO audio?

#24 opened 5 days ago by

dynamic vram not working in comfy

#23 opened 7 days ago by

please - do the same to TE

#22 opened 9 days ago by

Why is int8_convrot VAE slower than fp16 VAE?

#21 opened 9 days ago by

What is this used for?【minimax_h3_ref_lora_rank_256_bf16】

#20 opened 9 days ago by

Ref Lora Compatible Diffusion Model Question

#19 opened 9 days ago by

adding Lora freezes "Model Initializing ... "?

#18 opened 9 days ago by

Unprunned model versions w4a8 pLZ :)

#17 opened 10 days ago by

Un-pruned model versions w4a8 request

#16 opened 10 days ago by

w4a8 slower than int8??

#15 opened 10 days ago by

question about format and running in comfy

#14 opened 10 days ago by

minimax_h3_video_vae_int8_convrot.safetensors

#13 opened 10 days ago by

Custom node required?

#9 opened 12 days ago by

Update README.md

#8 opened 12 days ago by

KeyError: 'asym_w4a8_int8'

#7 opened 12 days ago by

VRAM savings?

#6 opened 12 days ago by

I found MiniMax-H3-Turbo-Lora ~ 4-step

#5 opened 12 days ago by

Great work! Please also make one for the ref2va version. Thanks!

#4 opened 12 days ago by

finally!

#3 opened 12 days ago by