---
license: mit
tags:
- onnx
- fp16
- windowsml
- onnxruntime
- directml
- video-restoration
- seedvr2
pipeline_tag: video-to-video
---

# SeedVR2-3B ONNX FP16

FP16 ONNX exports of **SeedVR2-3B** for WindowsML / ONNX Runtime.

This repository contains two ONNX layouts:

| Variant | Folder | Size | Use case |
|---|---:|---:|---|
| Monolithic | `monolithic/` | ~7.3 GB | Original ONNX export with one DiT step graph plus VAE encoder/decoder. |
| Block-sharded | `blocks/` | ~7.29 GB | FP16 block-sharded export for block-streaming execution and lower peak memory loaders. |

The repository provides model artifacts and metadata only. Use your own SeedVR2 inference pipeline or loader.

## Download

Download only the variant you need. Downloading the whole repository pulls both variants.

### Monolithic

```powershell
hf download Merserk/SeedVR2-3B-ONNX --include "monolithic/*" --include "config.json" --include "bucket_shapes.json" --include "versions*.json"
```

Python:

```python
from huggingface_hub import snapshot_download

snapshot_download(
    "Merserk/SeedVR2-3B-ONNX",
    allow_patterns=[
        "monolithic/*",
        "config.json",
        "bucket_shapes.json",
        "versions*.json",
    ],
)
```

### Block-sharded

```powershell
hf download Merserk/SeedVR2-3B-ONNX --include "blocks/*" --include "config.json" --include "bucket_shapes.json" --include "versions*.json"
```

Python:

```python
from huggingface_hub import snapshot_download

snapshot_download(
    "Merserk/SeedVR2-3B-ONNX",
    allow_patterns=[
        "blocks/*",
        "config.json",
        "bucket_shapes.json",
        "versions*.json",
    ],
)
```

## Files

### `monolithic/`

Keep each `.onnx` file together with its matching `.onnx.data` file:

```text
monolithic/
  seedvr2_step_cfg.onnx
  seedvr2_step_cfg.onnx.data
  seedvr2_vae_encoder.onnx
  seedvr2_vae_encoder.onnx.data
  seedvr2_vae_decoder.onnx
  seedvr2_vae_decoder.onnx.data
```

Default monolithic bucket:

```text
batch=1, frames=5, height=480, width=848
```

Width is 848 for the monolithic export metadata to preserve 16-pixel divisibility.

### `blocks/`

The block-sharded variant contains the DiT split into separate ONNX files:

```text
blocks/
  seedvr2_cfg_combine.onnx
  seedvr2_dit_embed.onnx
  seedvr2_dit_block_00.onnx
  ...
  seedvr2_dit_block_31.onnx
  seedvr2_dit_out.onnx
  seedvr2_vae_encoder.onnx
  seedvr2_vae_decoder.onnx
```

Primary block bucket:

```text
bucket_f1_256x256
batch=1, frames=1, height=256, width=256
```

For the primary block bucket, the VAE encoder accepts `[1,3,1,256,256]`, produces latent `[1,16,1,32,32]`, and the decoder returns `[1,3,1,256,256]`.

## Runtime notes

CUDA was used only for export acceleration. The ONNX artifacts target WindowsML-compatible ONNX Runtime execution. The block-sharded export is intended for DirectML block streaming, with CPU fallback possible for components where needed.

## Metadata

- `config.json` lists both variants and the exact paths inside this repository.
- `bucket_shapes.json` lists bucket shapes for both variants.
- `versions.json` points to variant-specific version metadata.
- `versions_monolithic.json` records the monolithic export metadata.
- `versions_block.json` records the block-sharded export metadata.

## Source

Source model: `ByteDance-Seed/SeedVR2-3B`.

This repository is an ONNX artifact export and is not the original SeedVR2 source repository.