#4925·sdnext

[Issue]: Cant generate video using HunyuanVideo 1.5 I2V

Author: normalfella112Created Jun 15, 2026Updated Sep 2, 2026

Labelscannot reproduce

Issue Description

Whenever I try to generate a video I see this, any ideas on how to fix it? Test log bellow.

Version Platform Description

test.txt

Relevant log output

bash

│E:\Anewforge\SDNEXT\sdnext\sdnext\venv\Lib\site-packages\diffusers\pipelines\hunyuan_video1_5\pipeline_hunyuan_video │
│                                                                                                                     │
│  783         # 4. Encode input prompt                                                                               │
│> 784         prompt_embeds, prompt_embeds_mask, prompt_embeds_2, prompt_embeds_mask_2 = self.encode_prompt(         │
│  785             prompt=prompt,                                                                                     │
ValueError: too many values to unpack (expected 4)

Operating system

Windows

Compute

nVidia CUDA

Interface

ModernUI

Branch

Master

Model

Unspecified

Acknowledgements

I have read the above and searched for existing issues
I confirm that this is classified correctly and its not an extension issue

Source: vladmandic/sdnext

View original on GitHub View discussion on GitHub