SeedVR2 (fastforks)

SeedVR2 (Wang et al., ByteDance Seed 2025): one-step video restoration and super-resolution by a diffusion transformer (NaDiT, adversarially post-trained) on the latents of a causal 3-D VAE. seedvr2-3b and seedvr2-7b are the 3 B and 7 B transformers (same VAE); frames are restored one at a time.

This repo is the model card only: it holds no weights. fastforks downloads the authors' weights from the links below and checks them against the SHA-256 in the model card. Code: https://github.com/kleinicke/fastforks. Kernels for the kernels backend: https://huggingface.co/fastforks/kernels. Original code: https://github.com/ByteDance-Seed/SeedVR at e4de8c24441a.

Author weight repos: https://huggingface.co/fastforks/seedvr2-3b-bf16, https://huggingface.co/fastforks/seedvr2-7b-bf16, https://huggingface.co/fastforks/seedvr2-vae-bf16.

Task: restoration. Tier: experimental. Outputs: image. Backends: portable, kernels.

Input: one RGB image. Output: the restored (upscaled) image, model.scale x the input size, RGB in [0, 1]. tile= runs large images in overlapping tiles (as ComfyUI).

Variants

Weights are the authors' files (not hosted here). SHA-256 shortened to 12 hex digits; the full hashes are in card.toml.

variant weights (author link) SHA-256 weights licence commercial use
seedvr2-3b@1 seedvr2-3b.safetensors
seedvr2-vae.safetensors
61df4f7979f4
bcf7f47d490e
Apache-2.0 yes
seedvr2-7b@1 seedvr2-7b.safetensors
seedvr2-vae.safetensors
bf73826841cf
bcf7f47d490e
Apache-2.0 yes

Licence

Original-code strings, recorded as they are (research-use package; check them before any other use). Code: Apache-2.0. Licence files: LICENSE, NOTICE.

Training data: Paper (arXiv 2506.05301): adversarial post-training of SeedVR on large-scale image and video data with synthetic degradations; the data sources and terms are not stated..

  • Licence note: The converted weight files (bf16 transformer with the fixed prompt embedding, bf16 VAE) are hosted by fastforks under the originals' Apache-2.0 licence; NOTICE keeps the attribution and lists the changes.

Licence evidence: https://github.com/ByteDance-Seed/SeedVR/blob/e4de8c24441a67e1b7df56abea10645059bb1185/LICENSE ; https://huggingface.co/ByteDance-Seed/SeedVR2-3B (license: apache-2.0)

Input size

any H x W >= 8, padded to a multiple of 8 Tested up to (portable, fp16, RTX 5060 Ti 16 GB): seedvr2-3b 512x512 (11.7 GiB peak allocated, 15.0 reserved); seedvr2-7b 512x512 (6.3 GiB peak allocated, 8.7 reserved)

Speed and memory

Restoration, synthetic, 512x512

model time (ms): original portable kernels kernels speed-up: vs original vs portable peak memory (MiB, allocated / reserved): original portable kernels
seedvr2-3b 13150 2295 2295 5.7x 1.0x 7598 / 15268 6140 / 6772 6140 / 6772

Startup and first call

model load (s): original portable kernels first call (s): original portable kernels
seedvr2-3b 96.20 4.88 4.90 13.89 2.49 2.50

torch.compile and TensorRT

model input time: portable compile TensorRT fp16 kernels preparation: compile (cold first call) ONNX + engine build peak MiB (alloc / reserved): portable compile kernels TensorRT (engine + torch)
seedvr2-3b 512x512 2295 ms deferred (build >5 min) fails: not run: onnx is not installed in the test venv 2295 ms – – 6140 / 6772 – 6140 / 6772 –
seedvr2-7b 512x512 – fails: load: OutOfMemoryError: CUDA out of memory. Tried to allocat deferred (build >5 min) – – – – – – –

Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB).

Hardware, protocol and caveats: benchmarks (RTX 5060 Ti 16 GB, sm_120). Original = the original code named above at the precision of portable (fp16, or (fp32) where fp16 is unsupported or not finite); original fp32 = its default fp32 run. Peak memory is the torch allocator's.

Accuracy

Parity with the original code

Every variant is checked against reference outputs of the original code at its pinned commit. CPU, float32: max |diff| <= 0.02 vs the original. GPU, bfloat16 (the default), portable and kernels: mean |diff| <= 0.01 and p99 <= 0.05 (absolute error of the output image).

Limitations

  • Generative restoration: the output depends on seed (one CPU generator draws the VAE posterior noise and the diffusion noise, upstream's default 666); other seeds give other plausible details.
  • One frame per call: the clip mode of the original (temporal attention over 4n + 1 frames) is not used by this image API.
  • The input is padded on the bottom and right (replicate) so that scale x its size is a multiple of 16, and the output cropped back; the original centre-crops the resized frame to a multiple of 16 instead.
  • No colour fix: the authors' scripts optionally match the output colours to the input with StableSR's wavelet reconstruction (non-commercial code), which is not included.
  • The 3 B model needs 7 GB for its weights plus the VAE's activations (14 GB peak at 960 x 1696 output); stream=False keeps them resident; by default (stream=True, CUDA) they stay in 6.6 GB of page-locked host RAM (9 GB peak at that size).
  • The 7 B model's weights are 16.5 GB in bf16: it runs with the default stream=True (blocks in 16.5 GB of page-locked host RAM; load fails with an error when less is free) or quant="nf4" (4.6 GB resident); with stream=false and no quant it runs out of GPU memory.
  • quant="nf4" changes the generated detail: 23-27 dB PSNR against the bf16 output (3 B, 4 clips; 7 B: 26-31 dB, 2 clips), about as far as another seed (18-24 dB), with ~3.5x smaller transformer blocks.

Usage

from fastforks import restoration

model = restoration.load("seedvr2-3b")  # mode="auto": the preferred backend from recorded measurements
result = model("photo.png")  # tile=512 for large images
result.image  # [3, sH, sW] float32 RGB in [0, 1], s = model.scale
result.backend  # what ran, e.g. Backend(mode='portable', device='cuda:0', dtype='float16')

Install with pip install fastforks (see https://github.com/kleinicke/fastforks). Other variants: seedvr2-3b, seedvr2-7b.

Kernels backend

restoration.load("seedvr2-3b", mode="kernels") runs our CUDA kernels (nf4). It needs a prebuilt kernel for your GPU architecture (downloaded from https://huggingface.co/fastforks/kernels on first use) and a recent NVIDIA driver; if there is none it raises BackendUnavailableError with the reason and never falls back silently. mode="auto" picks kernels when it can run and logs the choice.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Collection including fastforks/seedvr2