Workflows/Create video locally with HunyuanVideo 1.5

Video / WORKFLOW

Create video locally with HunyuanVideo 1.5

Use maintained ComfyUI templates for text-to-video or image-to-video. The task selector changes the exact checkpoint, input, files and steps.

Verified sourceSource checked 10/12/2026

LOCALRENTED GPU

Freshness 28 out of 100. Release age: 329 days old, Nov 20, 2025. Measures age, not quality.

Popularity 48 out of 100. 1,314 downloads in 30 days; 1,050 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.

Arena Open Source Text to video: rank 3 of 7. Ranked model: hunyuan-video-1.5. Original model ranking, not a benchmark of this package. Export dated Sep 22, 2026. Source checked Oct 15, 2026. Human preference ranking, not hardware suitability.

Category rankings & sources

Each position belongs to a separate Arena category. These are human preference rankings—not hardware recommendations. We do not combine the categories into one score.

Text to video#3of 7 open models

hunyuan-video-1.5tencentLicense: tencent-hunyuan-community

Original model ranking. This does not benchmark the selected quantization or package.

Export dated Sep 22, 2026Checked: Oct 15, 2026Overall board: #38

View official Text to video board

Image to video#2of 4 open models

hunyuan-video-1.5tencentLicense: tencent-hunyuan-community

Original model ranking. This does not benchmark the selected quantization or package.

Export dated Sep 22, 2026Checked: Oct 15, 2026Overall board: #40

View official Image to video board

Rankings reflect the dated export; live Arena boards can change. Open-source positions use Arena’s open-model filter. The overall position also includes the other models on that category board. Availability of weights and license terms still need to be checked separately.

What do these scores mean?

Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.

Freshness · release age

A new release starts at 100. The score halves every 180 days: 100 × 2^(−age in days / 180). When the release date is unavailable, a repository date is a fallback—not proof of when the model was released. Missing dates stay unscored.

Check the date source

Popularity · adoption

Downloads over the past 30 days carry 80% of the score; lifetime likes carry 20%. Both use a logarithmic scale, capped at 10 million downloads and 100,000 likes. If likes are unavailable, downloads carry all the weight. Missing download statistics stay unscored.

round(80 × min(1, log10(1 + downloads) / 7) + 20 × min(1, log10(1 + likes) / 5))

Check the statistics source

Scores as of Oct 15, 2026. Statistics checked Oct 15, 2026. Statistics are marked for refresh after 72 hours.Full scoring method

Before you begin

Hardware

The first run uses the maintained 720p templates. These are capacity starting points; render speed depends on frames and machine.

Machine	HunyuanVideo 1.5 · text to video	HunyuanVideo 1.5 · image to video
NVIDIA GPU · 16 GB VRAM	Below official 24 GB ComfyUI starting point; use RunPod.	Below official 24 GB starting point; use RunPod.
NVIDIA GPU · 24 GB VRAM	Publisher-supported starting class for five seconds at 720p.	Publisher-supported starting class; input image adds work.
NVIDIA GPU · 32 GB VRAM	More headroom; raise frames only after first clip.	More headroom; test the source image size.
AMD Radeon · 24 GB VRAM	Conditional ComfyUI backend; this guide has no checked ROCm route.	Conditional ComfyUI backend; use NVIDIA rental for this guide.
Intel Arc · 16 GB VRAM	No verified Hunyuan 1.5 route here; use NVIDIA rental.	No verified Hunyuan 1.5 route here; use NVIDIA rental.
Apple Silicon · 64 GB unified memory	No verified native route here; use NVIDIA rental.	No verified native route here; use NVIDIA rental.
CPU only · 64 GB RAM	Not a practical route in this guide; rent a GPU.	Not a practical route in this guide; rent a GPU.
RunPod · NVIDIA GPU · 24 GB VRAM	Good documented starting class; check storage.	Good documented starting class; check storage.

GPU fit is not a speed result. ComfyUI’s publisher guide supplies the 24 GB class; exact peak and runtime vary with template settings.

VIDEO · PROMPT OR IMAGESOURCE-LINKED SETUP

Turn one idea into a moving scene.

Choose text-to-video or image-to-video. The selection swaps the checkpoint, input steps, exact ComfyUI template, and file tree. Start with the five-second 720p setting.

Choose a package Load the visual workflow Check the result

BEFORE YOU BEGIN

Choose your package

Select the weights you want. The instructions, downloads and settings change with your choice.

Save your machine in My Hardware to get an automatic starting choice. You can always choose any package yourself.

CURRENT SETUP

HunyuanVideo 1.5 · text to video · official

Start with five seconds at 720p on a 24 GB NVIDIA GPU and leave disk/host-memory headroom. Increasing frames or upscaling costs more memory.

Freshness 28 out of 100. Release age: 329 days old, Nov 20, 2025. Measures age, not quality.

Popularity 48 out of 100. 1,314 downloads in 30 days; 1,050 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.

Category rankings & sources

Each position belongs to a separate Arena category. These are human preference rankings—not hardware recommendations. We do not combine the categories into one score.

Text to video#3of 7 open models

hunyuan-video-1.5tencentLicense: tencent-hunyuan-community

Original model ranking. This does not benchmark the selected quantization or package.

Export dated Sep 22, 2026Checked: Oct 15, 2026Overall board: #38

View official Text to video board

Image to video#2of 4 open models

hunyuan-video-1.5tencentLicense: tencent-hunyuan-community

Original model ranking. This does not benchmark the selected quantization or package.

Export dated Sep 22, 2026Checked: Oct 15, 2026Overall board: #40

View official Image to video board

What do these scores mean?

Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.

Freshness · release age

Check the date source

Popularity · adoption

round(80 × min(1, log10(1 + downloads) / 7) + 20 × min(1, log10(1 + likes) / 5))

Check the statistics source

Scores as of Oct 15, 2026. Statistics checked Oct 15, 2026. Statistics are marked for refresh after 72 hours.Full scoring method

View weight source

THE WORKFLOW

Set up HunyuanVideo 1.5 · text to video on your machine

Choose your computer first, then its GPU path. Follow the visual setup and import the linked workflow.

CHOOSE YOUR SETUPWindows

Other operating systems

Mac: The cited ComfyUI HunyuanVideo 1.5 guide gives a 24 GB NVIDIA GPU setup and does not document an Apple Silicon package or practical Mac capacity. Use the documented NVIDIA route locally or on RunPod. Compatibility source

Install ComfyUI

Download and update the Windows NVIDIA version of ComfyUI. If this template has red nodes, update ComfyUI before changing the graph.

Official ComfyUI download

Place every required file in its folder

Download the shared encoders and VAE plus the text-to-video checkpoint. Put each in the destination shown below; the optional 1080p upscaler is not needed for the first clip.

Put these files in ComfyUI/models/4 required files

text_encoders/

qwen_2.5_vl_7b_fp8_scaled.safetensorsShared prompt encoderGet file byt5_small_glyphxl_fp16.safetensorsShared text encoderGet file

vae/

hunyuanvideo15_vae_fp16.safetensorsShared video decoderGet file

diffusion_models/

hunyuanvideo1.5_720p_t2v_fp16.safetensorsText-to-video checkpointGet file

Open the exact Hunyuan template

Choose Hunyuan Video 1.5 T2V in the ComfyUI Template Library or import this JSON. Check the diffusion loader and every encoder against the filenames above.

Download the official ComfyUI JSON Read the official file tree and node guide

Enter a short video prompt

Select the T2V model, enter “A fox walks through a foggy forest as the camera slowly pans left,” and start at 1280×720 resolution with 121 frames (about five seconds at 24 fps). Keep the template sampler settings for the first run.

Queue, play and save

Click Queue. When generation finishes, play the video in the preview and find the saved file under ComfyUI/output. Check for actual motion, then save the prompt, seed, template name and package used. Lower frames or resolution if the GPU runs out of memory.

SUCCESS CHECK · HunyuanVideo 1.5 · text to video

A saved clip from your chosen input

Play the video under ComfyUI/output. Confirm motion follows the prompt, or that the selected image appears in the opening frames. Record which template, checkpoint and seed produced it.

ComfyUI publishes both importable JSON files. The optional upscaler is deliberately omitted from the first run; add it only after the base clip works. RunPod is an affiliate link.

Back to setup steps

When it doesn’t go to plan

ComfyUI shows red nodes or cannot find a checkpoint.

Update ComfyUI, import the matching T2V or I2V JSON, restart it, and compare every filename and folder with the official file tree.

The clip runs out of GPU memory.

Keep the first run at five seconds/720p, disable the optional super-resolution nodes, close other GPU apps, or select a larger rental GPU.

Image-to-video ignores the image.

Select the I2V package and matching diffusion checkpoint, then confirm the LoadImage and SigCLIP vision nodes use the source image and encoder.

REFERENCE LIBRARY

Sources and files

Original instructions, model files, and compatibility notes behind this setup.

2 SOURCES

01ComfyUI HunyuanVideo 1.5 templates, files and setupdocs.comfy.org 02Tencent checkpoint and community licensehuggingface.co

The model behind this workflow

HunyuanVideo 1.5

Keep exploring

Text

SETUP GUIDE

Qwen3.5 · 9B private chat setup

Source-backed package and platform-route plan for private local chat with Qwen3.5 · 9B. Preserves the official Ollama 9B package, a smaller 4B family fallback, and a clearly labeled community uncensored GGUF option.

Freshness 41 out of 100. Release age: 230 days old, Feb 27, 2026. Measures age, not quality.

Build a private coding assistant with Qwen3.8 27B

Run Qwen3.8 27B with Ollama, connect it to Cline in VS Code, and make one reviewable code change in a tiny local project.

Freshness 76 out of 100. Repository age: 71 days old, Aug 5, 2026. Measures age, not quality.

Arena Open Source Web coding: rank 10 of 55. Ranked model: qwen3.8-27b. Original model ranking, not a benchmark of this package. Export dated Oct 13, 2026. Source checked Oct 15, 2026. Human preference ranking, not hardware suitability.+3

SOFTWAREOllama + Visual Studio Code + Cline + Python

Create images locally with FLUX.2 Klein

Generate images from text prompts in ComfyUI with the fast distilled or more adjustable base 4B model. Save your results and reuse the visual workflow.

Freshness 35 out of 100. Release age: 273 days old, Jan 15, 2026. Measures age, not quality.

SOFTWAREComfyUI

LOCALRENTED GPU

Verified source