kijai/ComfyUI-HunyuanVideoWrapper

Issues

• HyVideoTextImageEncode Invalid image type. Expected either PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray, but got <class 'str'>.

  #223 opened 6 hours ago
  3

• vae error

  #222 opened 6 hours ago
  1

• DownloadAndLoadHyVideoTextEncoder

  #221 opened 6 hours ago
  0

• [Feat] mac MPS Support for fast version

  #220 opened 6 hours ago
  0

• Missing Nodes for VHE_VideoCombine

  #219 opened 6 hours ago
  0

• Hunyuan speed up by teacache

  #218 opened 6 hours ago
  0

• gguf compatability?

  #217 opened 6 hours ago
  0

• sombody use another model as text encoder, what's different from what you've introduced,

  #216 opened 6 hours ago
  1

• Non-Issue - Question on Prompt

  #215 opened a day ago
  0

• QUESTION: Max number of frames

  #214 opened a day ago
  0

• ip2v_example flow fails

  #213 opened 2 days ago
  2

• Using on Linux

  #212 opened 2 days ago
  1

• Could not allocate tensor with XXXXXXXX bytes. There is not enough GPU video memory available!

  #211 opened 2 days ago
  0

• cannot import name 'shard_checkpoint' from 'transformers.modeling_utils'

  #210 opened 3 days ago
  1

• VRAM Management

  #209 opened 3 days ago
  4

• "Replication_pad3d_cuda" not implemented for BFloat16

  #208 opened 3 days ago
  7

• i download the tencent's checkpoint,how can i convert the model to be used in this repo,i didn't wan to download twice.

  #207 opened 3 days ago
  3

• Question about quantization + others

  #206 opened 3 days ago
  2

• multiple Lora ussuie

  #205 opened 2 days ago
  0

• How to get better result from LoRA?

  #202 opened 3 days ago
  1

• Enhance a video giving very noisy results

  #201 opened 3 days ago
  5

• bad quality using fp8 model

  #200 opened 4 days ago
  3

• please

  #199 opened 4 days ago
  3

• Arcane lora?

  #198 opened 4 days ago
  1

• portage enhance video for workflow that supports GGuf models

  #197 opened 4 days ago
  0

• Invalid image type. Expected either PIL.Image

  #196 opened 4 days ago
  3

• Video2Video Example - Load Video and Frame Load

  #195 opened 5 days ago
  0

• Is it possible to use comfyui llava_llama3_fp8_scaled.safetensors for LLM text encoder

  #194 opened 5 days ago
  2

• STG Worth using?

  #193 opened 5 days ago
  3

• Support for GGUF versions?

  #192 opened 5 days ago
  6

• speed

  #191 opened 5 days ago
  3

• Problem with running the example

  #190 opened 5 days ago
  2

• Maximum frames/steps etc for 24GB card? Keep getting OOM

  #189 opened 6 days ago
  11

• Only vision_languague models support image input - HyVideoTextImageEncode

  #188 opened 6 days ago
  10

• About training HunyuanVideo

  #187 opened 6 days ago
  2

• Support of official VAE checkpoint

  #186 opened 6 days ago
  0

• Image Encode Error

  #185 opened 6 days ago
  2

• autodownload of llama failing

  #183 opened 6 days ago
  0

• using fast video on rtx 3090 getting out of memory

  #182 opened 6 days ago
  14

• Regarding the issue of the HunyuanVideo TextEncode node not matching the node demonstrated in the presentation.

  #181 opened 7 days ago
  2

• !!! Exception during processing !!! "replication_pad3d_cuda" not implemented for 'BFloat16'

  #180 opened 7 days ago
  8

• Can we support the official FP8 quantification model

  #179 opened 7 days ago
  3

• FP8 model quantization

  #178 opened 7 days ago
  0

• Optimization as comfyui official hunyuan workflow?

  #177 opened 7 days ago
  8

• Pointer argument (at 1) cannot be accessed from Triton (cpu tensor?)

  #176 opened 7 days ago
  0

• [BUG] Only one TextImageEncode allowed

  #175 opened 7 days ago
  4

• Is there a way to use the city96 GGUFs with this wrapper?

  #174 opened 7 days ago
  2

• Integrating FastVideo?

  #173 opened 7 days ago
  4

• README is a bit unclear given new changes

  #172 opened 7 days ago
  0

• redundant llava_lama copies

  #171 opened 7 days ago
  3