RAM 16GB and VRAM 12GB can run this workflow.

I hope someone can perform GGUF quantization on this model, as my hardware cannot handle it.

You can find the model at: https://huggingface.co/xiaozaa/catvton-flux-beta

Please put the merged model file in the ComfyUI/models/diffusion_models/ folder.

python
from safetensors.torch import load_file, save_file

# List of shard files
shards = [
    "diffusion_pytorch_model-00001-of-00003.safetensors",
    "diffusion_pytorch_model-00002-of-00003.safetensors",
    "diffusion_pytorch_model-00003-of-00003.safetensors"
]

# Merge models
merged = {}
for shard in shards:
    print(f"Loading shard: {shard}")
    merged.update(load_file(shard))

# Save merged result
save_file(merged, "merged_model.safetensors")
print("Merge completed: merged_model.safetensors")