Comfy-Org
/

MiniMax-H3

Model card Files Files and versions

Resources

View closed (8)

Why MiniMax H3 Ruins Faces on Wide Shots?

#30 opened about 6 hours ago by

Memory Error

#29 opened about 7 hours ago by

Fail forbidden trying to download minimax_h3_ref2va_pruned_fp8_scaled.safetensors

#28 opened about 12 hours ago by

Nvfp4 Pruned model for first last frame and reference to video

#27 opened about 12 hours ago by

Just a quick benchmark

#26 opened about 16 hours ago by

Minimax reference to video

#25 opened 1 day ago by

Not supported for Apple silicon

#24 opened 1 day ago by

Could you release an FP32 vae?

#23 opened 1 day ago by

Please release an INT4-quantized version of MiniMax-H3

#22 opened 1 day ago by

makisekurisu-jp

Update README.md

#21 opened 1 day ago by

Approximating VAE? (Or video preview in general)

#20 opened 1 day ago by

Upscaling 10s or 15s video to 2K failed in ref2va workflows

#19 opened 2 days ago by

[Help] RX 9070 XT (AMD) 16g vram + 5600x 32g + win11 – CUDA invalid argument when loading MiniMaxH3VideoVAE / any success stories?

#17 opened 2 days ago by

int4 convrot text encoder?

#16 opened 2 days ago by

Well, just in case anyone is wondering if can do i2v + ref, the answer is yes.

#15 opened 2 days ago by

Using minimax text encoder forTextGenerate node

#14 opened 2 days ago by

Prompt injected into every output file

#13 opened 2 days ago by

Where is the Notice file?

#12 opened 2 days ago by

Speedup 4-step LoRa ?

#11 opened 2 days ago by

what’s for pruned?

#10 opened 2 days ago by

the only thing i can say , LTX2.3 is like trash next to H3

#8 opened 2 days ago by

How much GPU memory is required for BF16?

#6 opened 2 days ago by

Will there be FP8 and NVFP4 of the models?

#5 opened 3 days ago by

VivekThePreceptor