Model
Trends
.ai
Model
Trends
.ai
the best open source ai models

EmptySD3LatentImage node in ComfyUI

EmptySD3LatentImage creates the 16-channel empty latent that Flux, SD3 and SD3.5 need; it replaces Empty Latent Image in those workflows.
Node pack: ComfyUI core
Category: Latent

What EmptySD3LatentImage does

SD3 and Flux use a VAE with 16 latent channels instead of the 4 channels of SD 1.5 and SDXL. A standard Empty Latent Image hands the sampler the wrong shape, which ends in noise or a size error. EmptySD3LatentImage produces the right tensor and otherwise behaves the same: width, height and batch_size in pixels.
Flux handles a wide range of sizes well, from 512 up to about 2 megapixels in one pass (1024x1024, 1152x896, 1344x768, 1536x1024). SD3.5 prefers 1024-class sizes.

Inputs

Name
Type
What it is
width
INT
Image width in pixels, multiple of 16 recommended.
height
INT
Image height in pixels.
batch_size
INT
Images per run.

Outputs

Name
Type
What it is
LATENT
LATENT
The empty 16-channel latent for KSampler.
AD
Skip the setup: ComfyUI workflows preinstalled on the cloud
BitVector runs ready-made ComfyUI workflows on its own GPUs. No install, no missing nodes, no red boxes. Open it on your phone or laptop and generate in a minute.

How to use EmptySD3LatentImage

  1. Add EmptySD3LatentImage instead of Empty Latent Image.
  2. Set 1024x1024 (or another Flux size).
  3. Connect to KSampler latent_image, denoise 1.0.

Settings and tips

  • Flux Dev makes good 1536x1024 images directly; no hires pass needed for many uses.
  • Multiples of 16 avoid edge artifacts with Flux.
  • Qwen-Image and HiDream also use this node.

Troubleshooting EmptySD3LatentImage

Flux output is noise although everything else is right

Why it happens
A 4-channel Empty Latent Image is wired into the sampler.

How to fix it
Swap it for EmptySD3LatentImage.

RuntimeError about channel sizes (expected 16, got 4)

Why it happens
Same cause: the latent has the wrong channel count for the model.

How to fix it
Use this node for Flux/SD3 and Empty Latent Image for SD 1.5/SDXL; also check that the VAE Encode in img2img uses the Flux VAE.

AD
Run this workflow from your phone
Every workflow on this page is preinstalled on BitVector with the models and custom nodes already in place. Pick one, type a prompt, done. Zero setup, nothing to download.

Questions about EmptySD3LatentImage

Does it work for Flux Schnell?

Yes, every Flux variant uses it.

Can I use it for SDXL?

No, SDXL needs the 4-channel Empty Latent Image.

Related nodes

Empty Latent Image
ComfyUI core
Empty Latent Image creates the blank latent canvas (width, height, batch_size) that text-to-image sampling starts from; dimensions must be multiples of 8 and match the model family.
Load Diffusion Model
ComfyUI core
Load Diffusion Model (UNETLoader) loads a bare denoising network such as Flux, SD3.5, Wan or HunyuanVideo from models/diffusion_models, with a weight_dtype option for fp8 to save VRAM.
Flux Guidance
ComfyUI core
Flux Guidance (FluxGuidance) writes the guidance value into the conditioning that Flux Dev was distilled to expect, replacing cfg; 3.5 is the default and 2 to 5 the useful range.
KSampler
ComfyUI core
KSampler runs the denoising loop: it takes the model, prompts and a latent and produces the finished latent image, controlled by seed, steps, cfg, sampler, scheduler and denoise.
AD
Tired of fixing nodes? Let the cloud do it
BitVector keeps hundreds of ComfyUI workflows installed, updated and tested on fast cloud GPUs. No Python, no CUDA errors, no VRAM limits. Works from any browser.

More ComfyUI nodes

CLIP Text Encode (Prompt)
ComfyUI core
CLIP Text Encode turns a text prompt into CONDITIONING using the model text encoder. One node holds the positive prompt, a second one the negative prompt.
ComfyUI Manager
ComfyUI-Manager
ComfyUI Manager is the extension that installs, updates and fixes custom node packs and models from inside the interface, resolves missing nodes in imported workflows and snapshots your setup.
Load Checkpoint
ComfyUI core
Load Checkpoint (CheckpointLoaderSimple) opens a .safetensors or .ckpt model file and hands out the three parts every workflow needs: the diffusion MODEL, the CLIP text encoder and the VAE.
Save Image
ComfyUI core
Save Image writes the IMAGE tensor to ComfyUI/output as a PNG, with the whole workflow embedded in the file metadata so the picture can be dragged back into ComfyUI to restore the graph.
VAE Decode
ComfyUI core
VAE Decode converts the sampled LATENT into a pixel IMAGE with the VAE; VAE Decode (Tiled) does the same in tiles for very large images.
Load Image
ComfyUI core
Load Image reads a picture from the ComfyUI input folder (or an upload) and outputs an IMAGE tensor plus a MASK taken from the alpha channel, the starting point for img2img, inpainting, ControlNet and IP-Adapter graphs.