SeedVR2-1.4B: 6-Layer Distilled Upscaler With ComfyUI Support

ComfyUI Wikinews

A 1.4B, 6-layer distillation of SeedVR2-7B brings one-step 4x/8x upscaling to low-VRAM machines, with a drop-in ComfyUI custom node and ready-made workflow.

SeedVR2-1.4B is a 1.44B-parameter, 6-layer knowledge distillation of ByteDance-Seed/SeedVR2-7B (the sharp EMA variant). It is a one-step diffusion image upscaler that cuts the teacher down from 8.2B parameters / 15.35 GB to 1.44B parameters / 2.69 GB, with a full ComfyUI support folder so the distilled architecture loads out of the box.

SeedVR2-1.4B 4x upscale comparison vs SeedVR2-3B

Why Distill to Six Layers

SeedVR2-7B's 36 transformer blocks are too heavy for many machines: 15.35 GB of fp16 weights and a 14–16 GB peak put it out of reach on consumer laptops, integrated GPUs, and phones. The 1.4B student keeps the same one-step diffusion formulation but uses only 6 transformer blocks, which cuts both the resident weights and the activation working set.

SeedVR2-1.4BSeedVR2-7B (teacher)Advantage
Transformer layers6366× fewer
Parameters1,442,608,2528,239,608,2325.7× fewer
Weights on disk (fp16)2.69 GB15.35 GB5.7× smaller
512→2048 (4×)20.2–22.5 s33.0–38.4 s≈1.6× faster
512→2048 peak RAM4.6 GB14.2–16.4 GB≈3× lighter
512→4096 (8×)54.4–55.2 s257–307 s4.7–5.6× faster

Measured on an Apple M2 Ultra (64 GB) via MLX, model load included.

At 4× the student tracks the 36-layer teacher closely on structure and actually delivers more edge energy: outputs read crisp and detailed rather than soft. If you are memory-bound, latency-bound, or batching a lot of images, this is the model to reach for.

ComfyUI Support (Drop-In Folder)

ComfyUI supports SeedVR2 natively, but it identifies the architecture by inspecting the checkpoint and only recognizes blocks.31 (3B) and blocks.35 (7B). The 1.4B model has six blocks, so it will not load until ComfyUI is taught about it. The repo ships a drop-in support folder that fixes this without editing any ComfyUI source files:

  1. Copy comfyui/ComfyUI-SeedVR2-1.4B/ into ComfyUI/custom_nodes/ and restart
  2. Use the ComfyUI variant checkpoint: seedvr2_distill_6L_1.4B_sharp_fp16_comfyui.safetensors (2.69 GB); it carries the positive_conditioning / negative_conditioning tensors ComfyUI's loader requires
  3. Pair it with the SeedVR2 VAE (ema_vae_fp16.safetensors, 0.47 GB)
  4. Drag in the ready-made workflow seedvr2_1.4b_upscale_image.json: Load Image → Upscale Image (SeedVR2 1.4B) → Save Image, one step, prompt-free
The ready-made ComfyUI workflow for SeedVR2-1.4B

Optional Distilled Decoder

The repo also includes an optional distilled VAE decoder (ema_vae_distill_fp16.safetensors, 0.19 GB) that is 112× smaller and ~13× faster to decode, at the cost of a mild low-pass effect. Swap it in via the ComfyUI-SeedVR2-1.4B-TinyVAE folder if you want the fastest possible decode.

Availability

Comments

Sign in with GitHub to join the discussion.

Loading comments…
SeedVR2-1.4B: 6-Layer Distilled Upscaler With ComfyUI Support | ComfyUI Wiki