Video models · Tencent

HY-OmniWeaving (8.3B, unified video)

Name: HY-OmniWeaving (8.3B, unified video)
Author: Tencent

Backbone: 8.3B
Native output: 848×480, 81 frames
Default steps: 50
License: tencent-hunyuan-community
Released: 2026-04

How to run it

Original weights · Hugging Face

Full-precision files from Tencent, for fine-tuning, vLLM, Transformers, Diffusers or ComfyUI.

huggingface.co/tencent/HY-OmniWeaving (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Tencent. The weights themselves download from Hugging Face.

github.com/Tencent-Hunyuan/OmniWeaving (opens in a new tab)

git clone https://github.com/Tencent-Hunyuan/OmniWeaving

Links checked 5 days ago against Hugging Face, Ollama and GitHub.

Apps that run HY-OmniWeaving (8.3B, unified video)

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or device in the Model Finder for an exact answer.

Mac (Apple silicon)

Draw Things

Native Mac app with its own fast Metal engine and quantized downloads.

Also good:

Windows PC, NVIDIA GPU

ComfyUI

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Also good:

Linux, NVIDIA or AMD GPU

ComfyUI

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Also good:

App	Type	Opens it as	Systems	Engine	Start
ComfyUI	Desktop app	Built-in support	macOSWindowsLinux	PyTorch, GGUF via ComfyUI-GGUF	Get ComfyUI
Draw Things	Desktop app	Built-in support	macOSiOS	Draw Things	Open in Draw Things
SwarmUI	Desktop app	Built-in support	macOSWindowsLinux	ComfyUI	Get SwarmUI
SD.Next	Desktop app	Built-in support	macOSWindowsLinux	PyTorch	Get SD.Next
Stability Matrix	Desktop app	Built-in support	macOSWindowsLinux	installs ComfyUI, Forge, InvokeAI, SwarmUI, SD.Next or Fooocus	Get Stability Matrix
WanGP	Desktop app	Built-in support	WindowsLinux	PyTorch	Get WanGP

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Peak GPU memory at the native output size. Offloading moves the text encoder, or the whole pipeline, to system RAM between steps.

Precision	Everything on GPU	Text encoder offloaded	Sequential offload	Download
`FP16`	41.9 GB	24.1 GB	10.7 GB	Original weights (opens in a new tab)
`FP8`	24.6 GB	15.7 GB	7.8 GB	Original weights (opens in a new tab)

Components

Backbone: 8.34B
VAE: 1.26B
Qwen2.5-VL 7B (OmniWeaving fine-tune): 8.29B
ByT5: 0.2B
SigLIP (image): 0.43B

Best hardware for HY-OmniWeaving (8.3B, unified video)

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc B570 (launch price $219) at FP8.

Quality	Hardware	Quant	Memory	Speed	Runs
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5090	`FP8`	24.6 GB / 32.0 GB	6.7 min	Runs well
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell	`FP16`	41.9 GB / 96.0 GB	6.5 min	Runs well
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4090	`FP8`	15.7 GB / 24.0 GB	8.3 min	Runs wellencoder in RAM
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon RX 7900 XTX	`FP8`	15.7 GB / 24.0 GB	9.3 min	Runs wellencoder in RAM
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX 5000 Ada Generation	`FP8`	24.6 GB / 32.0 GB	9.5 min	Runs well
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 80GB SXM	`FP16`	41.9 GB / 80.0 GB	6.2 min	Runs well
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H800 80GB SXM	`FP16`	41.9 GB / 80.0 GB	6.2 min	Runs well
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA A100 40GB	`FP8`	24.6 GB / 40.0 GB	8.3 min	Runs well
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon AI PRO R9700	`FP8`	24.6 GB / 32.0 GB	10.7 min	Runs well
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon PRO W7800 32GB	`FP8`	24.6 GB / 32.0 GB	11.1 min	Runs well

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.