Black latents and black rendered video

Nobody has claimed this yet.

Hi, i am experiencing a black render issue with FramePack !!

I have artx 4080 super with 16go Vram, python 3.10.6, Torch 2.6.0 + cu126, Torchvision 0.21.0 + cu126
There is no error message, the generation go to it's end. The latents are black and also the rendered video!

**Command log:**

Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Loaded CLIPTextModel to cuda:0 as complete.
Unloaded CLIPTextModel as complete.
Loaded AutoencoderKLHunyuanVideo to cuda:0 as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Loaded SiglipVisionModel to cuda:0 as complete.
latent_padding_size = 27, is_last_section = False
Unloaded SiglipVisionModel as complete.
Moving DynamicSwap_HunyuanVideoTransformer3DModelPacked to cuda:0 with preserved memory: 6 GB
0%| | 0/25 [00:00

Start at webui/demo_gradio.py:227, where the runtime warning occurs during preview conversion, and reproduce with the reported Python, Torch, Torchvision, and CUDA versions. Trace the generation and decoding logs for invalid values in the latents and rendered video. Done means generation completes without invalid-value warnings and produces non-black latents and video.

Written by the indexing model from the issue text.