FastWan2.2-TI2V-5B-FullAttn-Diffusers by FastVideo

Runs On

What it takes to serve FastWan2.2-TI2V-5B-FullAttn-Diffusers (5B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

Precision	Weights	Memory needed	Cheapest setup	Per hour	Also fits
16-bit	10.0 GB	12.0 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
8-bit	5.0 GB	6.0 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
4-bit	2.5 GB	3.0 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

By FastVideo, published under apache-2.0, revision 3e187042a324.

FastVideo FastWan2.2-TI2V-5B-FullAttn-Diffusers Model

FastVideo Team

HF Paper (VSA) | arXiv Paper (VSA) | Github | Project Page

Online Demo

You can try our models here！

Introduction

We're excited to introduce the FastWan2.2 series—a new line of models finetuned with our novel Sparse-distill strategy. This approach jointly integrates DMD and VSA in a single training process, combining the benefits of both distillation to shorten diffusion steps and sparse attention to reduce attention computations, enabling even faster video generation.

FastWan2.2-TI2V-5B-Full-Diffusers is built upon Wan-AI/Wan2.2-TI2V-5B-Diffusers. It supports efficient 3-step inference and produces high-quality videos at 121×704×1280 resolution. For training, we used simulated forward for the generator model, making the process data-free. The current FastWan2.2-TI2V-5B-Full-Diffusers model is trained using only DMD.

Model Overview

Read the full model card (413 words)

Identity and Version

Repository: FastVideo/FastWan2.2-TI2V-5B-FullAttn-Diffusers
Publisher: FastVideo
Task: Text to video
Modality: Video
Library: diffusers
Parameters: 5B parameters
Languages: Not stated by the source
Revision: 3e187042a324f6f5fb68fd22110a78725253de8f
First published: 2025-08-02
Last updated: 2025-11-25

Files and Weights

25 files, 24.2 GB in total. The weights are 5 files totalling 24.2 GB in safetensors.

Weights5 files · 24.2 GB

Configuration7 files · 33.9 KB

Tokenizer3 files · 21.4 MB

Documentation1 file · 4.3 KB

Other8 files · 1.6 MB

Repository1 file · 1.9 KB

Every file

File	Type	Size	SHA-256
text_encoder/model-00001-of-00003.safetensors	Weights	4.9 GB	a8e861969c74
text_encoder/model-00002-of-00003.safetensors	Weights	5.0 GB	d57d948ece48
text_encoder/model-00003-of-00003.safetensors	Weights	1.4 GB	0da9ee284e21
transformer/diffusion_pytorch_model.safetensors	Weights	10.0 GB	a1cf4acd5d02
vae/diffusion_pytorch_model.safetensors	Weights	2.8 GB	62cd18f19438
model_index.json	Configuration	502 B	—
scheduler/scheduler_config.json	Configuration	820 B	—
text_encoder/config.json	Configuration	855 B	—
text_encoder/model.safetensors.index.json	Configuration	22.5 KB	—
tokenizer/special_tokens_map.json	Configuration	7.1 KB	—
transformer/config.json	Configuration	503 B	—
vae/config.json	Configuration	1.7 KB	—
README.md	Documentation	4.3 KB	—
assets/comp_effic.png	Other	202.2 KB	75ee012dcfb0
assets/logo.png	Other	56.3 KB	—
assets/moe_2.png	Other	527.9 KB	4ea471ccb643
assets/moe_arch.png	Other	74.9 KB	—
assets/performance.png	Other	306.5 KB	97ef99c13c8a
assets/vae.png	Other	165.5 KB	4aaea5e187f1
examples/i2v_input.JPG	Other	250.6 KB	077e3d965090
test.txt	Other		—
.gitattributes	Repository	1.9 KB	—
tokenizer/spiece.model	Tokenizer	4.5 MB	e3909a67b780
tokenizer/tokenizer.json	Tokenizer	16.8 MB	20a46ac25674
tokenizer/tokenizer_config.json	Tokenizer	61.8 KB	—

License and Download

License: apache-2.0
Access: Open weights, no gate
Download size: 24.2 GB

Download from FastVideo

Released by FastVideo through its official repository on Hugging Face. Read the license.

Built From

Described by arXiv:2502.04507
Described by arXiv:2505.13389

Memory Requirements

Precision	Weights in memory
As published	24.2 GB
16-bit	10.0 GB
8-bit	5.0 GB
4-bit	2.5 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About FastWan2.2-TI2V-5B-FullAttn-Diffusers

How much GPU memory does FastWan2.2-TI2V-5B-FullAttn-Diffusers need?

About 12 GB at 16-bit and 3 GB at 4-bit: the weights (5B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run FastWan2.2-TI2V-5B-FullAttn-Diffusers on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use FastWan2.2-TI2V-5B-FullAttn-Diffusers commercially?

Yes. FastWan2.2-TI2V-5B-FullAttn-Diffusers is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

Model · Text to video

Wan2.2-TI2V-5B-Diffusers

Wan-AI

We are excited to introduce Wan2.2, a major upgrade to our foundational video models. With Wan2.2, we have focused on incorporating the following innovations: This repository contains our TI2V-5B model, built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research. Your browser does not support the video tag. If your research or project builds upon Wan2.1 or Wan2.2, we welcome you to share it…

Open weights apache-2.0 5B parameters diffusers

View model

Model · Text to video

LongLive-2.0-5B-Diffusers

Yixiao Zeng

A diffusers-directory-layout repackaging of Efficient-Large-Model/LongLive-2.0-5B so it loads directly in SGLang Diffusion (sglang.multimodalgen) without any runtime overlay/materialization. - transformer/ — the generator weights extracted from the original modelbf16.pt, kept in their original (model.) naming; SGLang's LongLive2Transformer3DModel.paramnamesmapping maps them to the diffusers module names at load (same convention as LingBot-World). - scheduler / textencoder / tokenizer / vae — taken from Wan-AI/Wan2.2-TI2V-5B-Diffusers. - modelindex.json classname = LongLive2Pipeline.

Open weights 5B parameters diffusers

View model

Model · Text to video

CogVideoX-2b

Z.ai

Visit QingYing and API Platform to experience commercial video generation models. CogVideoX is an open-source version of the video generation model originating from QingYing. The table below displays the list of video generation models we currently offer, along with their foundational information. Data Explanation + When testing using the diffusers library, all optimizations provided by the diffusers library were enabled. This solution has not been tested for actual VRAM/memory usage on devices other than NVIDIA A100 / H100. Generally, this solution can be adapted to all devices with NVIDIA Ampere architecture and above. If the optimizations are disabled, VRAM usage will increase…

Open weights apache-2.0 1.7B parameters diffusers

View model

Model · Text to video

TurboWan2.1-T2V-1.3B-Diffusers

PostYellow

convert TurboWan2.1-T2V-1.3B-480P(https://modelscope.cn/models/TurboDiffusion/TurboWan2.1-T2V-1.3B-480P/summary) to TurboWan2.1-T2V-1.3B-Diffusers convert script https://github.com/IPostYellow/TurboWantoDiffusers/blob/main/convertturbowantodiffusers.py To use in sglang

Open weights apache-2.0 1.4B parameters diffusers

View model

Model · Text to video

Wan2.1-T2V-1.3B-Diffusers

Wan-AI

In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features: This repository hosts our T2V-1.3B model, a versatile solution for video generation that is compatible with nearly all consumer-grade GPUs. In this way, we hope that Wan2.1 can serve as an easy-to-use tool for more creative teams in video creation, providing a high-quality foundational model for academic teams with limited computing resources. This will facilitate both the rapid development of the video creation community and the swift advancement of video technology. Your browser does not support the video tag.…

Open weights apache-2.0 1.4B parameters diffusers

View model

Model · Text to video

Wan2.1-T2V-1.3B

Wan-AI

In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features: This repository hosts our T2V-1.3B model, a versatile solution for video generation that is compatible with nearly all consumer-grade GPUs. In this way, we hope that Wan2.1 can serve as an easy-to-use tool for more creative teams in video creation, providing a high-quality foundational model for academic teams with limited computing resources. This will facilitate both the rapid development of the video creation community and the swift advancement of video technology. Your browser does not support the video tag.…

Open weights apache-2.0 1.4B parameters diffusers

View model