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

Save Image node in ComfyUI

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.
Node pack: ComfyUI core
Category: Image

What Save Image does

Save Image is the last node of most graphs. It converts the IMAGE tensor to 8-bit PNG, writes it to the output folder (optionally a subfolder) with an automatic counter, and embeds the prompt and the full workflow JSON in the PNG text chunks. Dropping such a PNG on the ComfyUI canvas later rebuilds the entire workflow, which is how most shared workflows travel.
filename_prefix can include a folder (renders/portrait), and tokens such as %date:yyyy-MM-dd% or %Empty Latent Image.width% to pull values from other nodes into the name. Images in a batch are saved as separate files with _00001_, _00002_ suffixes. For a quick look without writing to disk, use Preview Image instead.

Inputs

Name
Type
What it is
images
IMAGE
The pictures to save (one or many in a batch).
filename_prefix
STRING
Name prefix, may contain a subfolder and %tokens%.

Outputs

Name
Type
What it is
(none)
-
Output node: shows the saved pictures in the node and writes them to ComfyUI/output.
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 Save Image

  1. Connect the IMAGE from VAE Decode (or an upscaler) to images.
  2. Set filename_prefix, for example renders/%date:yyyy-MM-dd%/flux.
  3. Queue the prompt; the pictures appear in the node and in ComfyUI/output.
  4. Drag a saved PNG back into ComfyUI to restore the workflow.
  5. Use the Queue > History panel to find previous outputs.

Settings and tips

  • Start ComfyUI with --output-directory to save somewhere else, for example a synced folder.
  • --disable-metadata removes the embedded workflow when you share images publicly.
  • The counter never overwrites: deleting files leaves gaps, which is harmless.
  • For JPEG or WebP output use the Image Save node from WAS or the Save Image Extended pack; core saves PNG only.
  • Chain two Save Image nodes to keep both the base and the upscaled picture.

Troubleshooting Save Image

Images are not in the output folder

Why it happens
A filename_prefix with a subfolder, a custom --output-directory, or the node is actually Preview Image (which saves to temp).

How to fix it
Check the prefix for a slash and the start command for --output-directory. Preview Image files live in ComfyUI/temp and are deleted on restart.

PermissionError when saving

Why it happens
The output folder is read-only or on a disconnected network drive, common with portable installs run from a protected folder.

How to fix it
Move ComfyUI out of Program Files or pass --output-directory to a writable path.

Dragging the PNG into ComfyUI does not restore the workflow

Why it happens
Metadata was stripped: the server runs with --disable-metadata, or the file passed through a messenger or image host that recompressed it.

How to fix it
Share the original PNG from the output folder (zip it to prevent recompression) or export the workflow as JSON from the menu.

Colours are washed out compared with the preview

Why it happens
The viewer applies a different colour profile; PNGs from ComfyUI carry no ICC profile.

How to fix it
View in an sRGB-managed viewer; the data is identical to the preview.

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 Save Image

Can Save Image write JPEG?

Not the core node. Use Preview Image for throwaways, or an extended save node for JPEG/WebP with quality settings.

Where is the metadata stored?

In the PNG tEXt chunks "prompt" and "workflow". Any PNG metadata viewer shows them.

Can I save 16-bit PNGs?

No, output is 8-bit RGB. Custom save nodes offer 16-bit or EXR for compositing work.

Related nodes

Preview Image
ComfyUI core
Preview Image displays an IMAGE inside the node without saving it to the output folder (a temporary file is kept until restart), ideal for checking intermediate steps like masks, ControlNet maps and hires passes.
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.
Video Combine (VideoHelperSuite)
ComfyUI-VideoHelperSuite
Video Combine turns an IMAGE batch of frames into an MP4, WebM, GIF or image sequence at a chosen frame rate, with optional audio, loop count and ping-pong, the standard output node for every video workflow.
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.
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.
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.
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 LoRA
ComfyUI core
Load LoRA (LoraLoader) applies a LoRA file to the MODEL and CLIP with separate strengths, so a style, character or concept can be added to any checkpoint without merging.