HunyuanVideo

---

README(From Huggingface)

---

base_model:

• tencent/HunyuanVideo library_name: diffusers

---

Unofficial community fork for Diffusers-format weights on tencent/HunyuanVideo.

Using Diffusers

HunyuanVideo can be used directly from Diffusers. Install the latest version of Diffusers.

import paddle
from diffusers import HunyuanVideoPipeline, HunyuanVideoTransformer3DModel
from diffusers.utils import export_to_video

model_id = "hunyuanvideo-community/HunyuanVideo"
transformer = HunyuanVideoTransformer3DModel.from_pretrained(
    model_id, subfolder="transformer", dtype=paddle.bfloat16
)
pipe = HunyuanVideoPipeline.from_pretrained(model_id, transformer=transformer, dtype=paddle.float16)

# Enable memory savings
pipe.vae.enable_tiling()
pipe.enable_model_cpu_offload()

output = pipe(
    prompt="A cat walks on the grass, realistic",
    height=320,
    width=512,
    num_frames=61,
    num_inference_steps=30,
).frames[0]
export_to_video(output, "output.mp4", fps=15)

Refer to the documentation for more information.

Model Files

• README.md (1.1 KB)

• config.json (38.0 B)

• model_index.json (563.0 B)

• scheduler/scheduler_config.json (419.0 B)

• text_encoder/config.json (1.1 KB)

• text_encoder/model_state-00001-of-00002.pdparams (9.3 GB)

• text_encoder/model_state-00002-of-00002.pdparams (4.7 GB)

• text_encoder/model_state.pdparams.index.json (22.5 KB)

• text_encoder_2/config.json (4.4 KB)

• text_encoder_2/model_state.pdparams (1.6 GB)

• tokenizer/special_tokens_map.json (577.0 B)

• tokenizer/tokenizer.json (16.4 MB)

• tokenizer/tokenizer_config.json (50.5 KB)

• tokenizer_2/merges.txt (512.4 KB)

• tokenizer_2/special_tokens_map.json (478.0 B)

• tokenizer_2/tokenizer_config.json (194.0 B)

• tokenizer_2/vocab.json (842.1 KB)

• transformer/config.json (543.0 B)

• transformer/diffusion_paddle_model-00001-of-00003.safetensors (9.3 GB)

• transformer/diffusion_paddle_model-00002-of-00003.safetensors (9.3 GB)

• transformer/diffusion_paddle_model-00003-of-00003.safetensors (5.3 GB)

• transformer/diffusion_paddle_model.safetensors.index.json (128.8 KB)

• vae/config.json (751.0 B)

• vae/diffusion_paddle_model.safetensors (940.3 MB)

Back to Main