Model reference · open weights
HunyuanVideo-1.5-480p_t2v is an open-weight video model from hunyuanvideo-community. HunyuanVideo-1.5-Diffusers-480p_t2v (FP32) weighs 26.7 GB; the smallest configuration that runs it is RTX 4090 24 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v model card, 2026-10-01
What it is
| Released by | hunyuanvideo-community |
|---|---|
| Released | 2025-11-27 |
| Parameters | 8.3B |
| VRAM | 26.7 GB for the weights |
What it runs on
| Card | Weights | Memory |
|---|---|---|
| RTX 3060 12 GB … RTX 4060 Ti 16 GB | does not fit | |
| RTX 3090 24 GB | fits (encoders offloaded) | 23.4 GB |
| RTX 4090 24 GB | fits (encoders offloaded) | 23.4 GB |
| RTX 5090 32 GB | tight | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
Hunyuan1.5 use attention masks with variable-length sequences. For best performance, we recommend using an attention backend that handles padding efficiently.
We recommend installing kernels (pip install kernels) to access prebuilt attention kernels.
You can check our documentation to learn more about all the different attention backends we support.
import torch
dtype = torch.bfloat16
device = "cuda:0"
from diffusers import HunyuanVideo15Pipeline, attention_backend
from diffusers.utils import export_to_video
pipe = HunyuanVideo15Pipeline.from_pretrained("hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v", torch_dtype=dtype)
pipe.enable_model_cpu_offload()
pipe.vae.enable_tiling()
generator = torch.Generator(device=device).manual_seed(seed)
with attention_backend("_flash_3_hub"): # or '"flash_hub"' if you are not using H100/H800
video = pipe(
prompt=prompt,
generator=generator,
num_frames=121,
num_inference_steps=50,
).frames[0]
export_to_video(video, "output.mp4", fps=24)
Of course, you can still run inference with default attention backend.
import torch
dtype = torch.bfloat16
device = "cuda:0"
from diffusers import HunyuanVideo15Pipeline
from diffusers.utils import export_to_video
pipe = HunyuanVideo15Pipeline.from_pretrained("hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v", torch_dtype=dtype)
pipe.enable_model_cpu_offload()
pipe.vae.enable_tiling()
generator = torch.Generator(device=device).manual_seed(seed)
video = pipe(
prompt=prompt,
generator=generator,
num_frames=121,
num_inference_steps=50,
).frames[0]
export_to_video(video, "output.mp4", fps=24)
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
Running it yourself
Rent a machine by the hour. Runs as it is with diffusers, on the machine, in Python.
# on your rented machine: pip install diffusers transformers accelerate ftfy
import torch
from diffusers import DiffusionPipeline
from diffusers.utils import export_to_video
pipe = DiffusionPipeline.from_pretrained("hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v", torch_dtype=torch.bfloat16).to("cuda")
frames = pipe(prompt="a drone shot over a forest at sunrise").frames[0]
export_to_video(frames, "/workspace/out.mp4", fps=16)