Model reference · open weights
SeedVR2 is an open-weight video model from Johnmn, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
SeedVR2: One-Step Video Restoration via Diffusion Adversarial Post-Training Jianyi Wang, Shanchuan Lin, Zhijie Lin, Yuxi Ren, Meng Wei, Zongsheng Yue, Shangchen Zhou, Hao Chen, Yang Zhao, Ceyuan Yang, Xuefeng Xiao, Chen Change Loy, Lu Jiang src="https://img.shields.io/badge/SeedVR2-Website-0A66C2?logo=safari&logoColor=white" alt="SeedVR Website" / src="https://img.shields.io/badge/SeedVR2-Paper-red?logo=arxiv&logoColor=red" alt="SeedVR2 Paper on ArXiv" / alt="Github" src="https://img.shields.io/badge/SeedVR2-Codebase-536af5?color=536af5&logo=github" alt="SeedVR2 Codebase" / src="https://img.shields.io/badge/SeedVR-Models-yellow?logo=huggingface&logoColor=yellow" alt="SeedVR Models" / src="https://img.shields.io/badge/SeedVR2-Space-orange?logo=huggingface&logoColor=yellow" alt="SeedVR2 Space" / src="https://img.shields.io/badge/Demo%20Video-%23FF0000.svg?logo=YouTube&logoColor=white" alt="SeedVR2 Video Demo on YouTube" / Recent advances in diffusion-based video restoration (VR) demonstrate significant improvement in visual quality, yet yield a prohibitive computational cost during inference. While several distillation-based approaches have exhibited the potential of one-step image restoration, extending existing approaches to VR remains challenging and underexplored, due to the limited generation ability and poor temporal consistency, particularly when dealing with high-resolution video in real-world settings. In this work, we propose a one-step diffusion-based VR model, termed as SeedVR2, which performs adversarial VR training against real data. To handle the challenging high-resolution VR within a single step, we introduce several enhancements to both model architecture and training procedures. Specifically, an adaptive window attention mechanism is proposed, where the window size is dynamically adjusted to fit the output resolutions, avoiding window inconsistency observed under high-resolution VR using window attention with a predefined window size. To stabilize and improve the adversarial post-training towards VR, we further verify the effectiveness of a series of losses, including a proposed feature matching loss without significantly sacrificing training ef
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | Johnmn |
|---|---|
| Type | Video models |
| Variants | 1 |
| Runs with | seedvr |
| Released | 2026-04-21 |
| Popularity | 2k downloads / month |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| SeedVR2-3B | — | BF16 | — | — | Weights ↗ |
Using it via the API
Once AxForge deploys johnmn-seedvr2 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (johnmn-seedvr2 below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/videos/generations \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"johnmn-seedvr2","prompt":"a drone shot over a forest"}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore