SAVRN
Search Contact SAVRN

Open-weight model · Video to video

SwiftVR

by H Oliday H-oliday/SwiftVR

SwiftVR is an open-weight model for video to video from H Oliday, released under Apache License 2.0. It has 5B parameters. At 16-bit it needs about 12 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 969 downloads a month.

SwiftVR is a streaming one-step generative video restoration (VR) framework presented in SwiftVR: Real-Time One-Step Generative Video Restoration.

Parameters5B
Context
Weights20.2 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads969

Runs On

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

PrecisionWeightsMemory neededCheapest setupPer hourAlso 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 24, 2026.

SwiftVR on every accelerator the SAVRN Index prices, at every precision

Model Card

By H Oliday, published under apache-2.0, revision 5e89ae51d17b.

SwiftVR: Real-Time One-Step Generative Video Restoration

SwiftVR is the first generative video restoration model to reach real-time 1080p streaming on a consumer-grade GPU (≈26 FPS on a single RTX 5090), sustains 31 FPS at QHD (2560×1440) and 14 FPS at 4K (3840×2160) on a single H100, and streams at resolutions where every compared diffusion-based VR baseline runs out of memory.

SwiftVR is a streaming one-step generative video restoration (VR) framework presented in SwiftVR: Real-Time One-Step Generative Video Restoration.

Updates

  • [2026/06] Release the inference code and pretrained weights

Community Works

  • LightX2V brings faster inference and lower GPU memory usage to SwiftVR (1.91× speedup and 65.71% lower peak GPU memory per request on a single H100). It also supports multi-GPU acceleration. Ready-to-use scripts cover image and video super-resolution, offline inference, and API serving. Get started with LightX2V →

Highlights

Read the full model card (776 words)

Identity and Version

Repository
H-oliday/SwiftVR
Publisher
H Oliday
Task
Video to video
Modality
Other
Library
diffusers
Parameters
5B parameters
Languages
Not stated by the source
Revision
5e89ae51d17b564b40cb14029002b48a83c2ff50
First published
2026-06-02
Last updated
2026-09-24

Files and Weights

11 files, 20.3 GB in total. The weights are 3 files totalling 20.2 GB in safetensors.

Weights3 files · 20.2 GB
Configuration1 file · 495 B
Documentation1 file · 8.1 KB
Other5 files · 178.0 MB
Repository1 file · 438 B
Every file
FileTypeSizeSHA-256
prompt_embedding.safetensorsWeights4.2 MB cc4cf7b9aa9d
reae.safetensorsWeights163.8 MB c915205d1833
transformer/diffusion_pytorch_model.safetensorsWeights20.0 GB f7ade5b8f7f4
transformer/config.jsonConfiguration495 B
README.mdDocumentation8.1 KB
assets/demo_1.mp4Other47.9 MB a16afce37f24
assets/demo_2.mp4Other65.6 MB f70c43561aca
assets/demo_3.mp4Other60.5 MB 696127bbaf6f
assets/qualitative.pngOther2.8 MB 856b43645149
assets/teaser.avifOther1.2 MB cdbe7f2bfb72
.gitattributesRepository438 B

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
20.2 GB
Download from H Oliday

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

Built From

  • Described by arXiv:2606.09516

Memory Requirements

PrecisionWeights in memory
As published20.2 GB
16-bit10.0 GB
8-bit5.0 GB
4-bit2.5 GB

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

Questions About SwiftVR

How much GPU memory does SwiftVR 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 SwiftVR 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 SwiftVR commercially?

Yes. SwiftVR 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.