298977887/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2025-01-11 02:15:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Adjust used dtypes for hunyuan video VAE and diffusion model.

Browse Source
This commit is contained in:
comfyanonymous 2024-12-16 23:31:10 -05:00
parent 0b25f47bd9
commit 39b1fc4ccc
2 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
comfy/sd.py
View File

@ -323,6 +323,7 @@ class VAE:
self.first_stage_model = AutoencoderKL(ddconfig=ddconfig, embed_dim=sd['post_quant_conv.weight'].shape[1])
self.memory_used_decode = lambda shape, dtype: (1500 * shape[2] * shape[3] * shape[4] * (4 * 8 * 8)) * model_management.dtype_size(dtype)
self.memory_used_encode = lambda shape, dtype: (900 * max(shape[2], 2) * shape[3] * shape[4]) * model_management.dtype_size(dtype)
self.working_dtypes = [torch.bfloat16, torch.float16, torch.float32]
elif "decoder.layers.1.layers.0.beta" in sd:
self.first_stage_model = AudioOobleckVAE()

2
comfy/supported_models.py
View File

@ -753,7 +753,7 @@ class HunyuanVideo(supported_models_base.BASE):
memory_usage_factor = 2.0 #TODO
supported_inference_dtypes = [torch.bfloat16, torch.float16, torch.float32]
supported_inference_dtypes = [torch.bfloat16, torch.float32]
vae_key_prefix = ["vae."]
text_encoder_key_prefix = ["text_encoders."]