How to Use FFmpeg xstack with Different Video Dimensions

This guide explains how to use FFmpeg’s xstack filter to combine and stack multiple videos that have different resolutions and dimensions. You will learn the core concepts of the xstack layout grid, how to position videos of varying sizes using precise coordinate positioning, and how to use scaling or padding to ensure a clean visual output without rendering artifacts.

Understanding the xstack Layout System

The xstack filter positions multiple input videos on a single output canvas using a 2D coordinate system (X_Y). It uses variables to represent the dimensions of the inputs: * w0, h0 represent the width and height of the first input. * w1, h1 represent the width and height of the second input, and so on.

When stacking videos of different sizes, you must explicitly define these coordinates in the layout parameter and handle any empty spaces created by the size mismatches.

Method 1: Direct Stacking with Custom Coordinates

If you want to keep the original resolutions of your videos, you can position them side-by-side or stacked vertically using their exact dimensions.

Example: Side-by-Side (Horizontal Stack)

Suppose you have two videos: * Video 0: 1280x720 * Video 1: 640x360

To stack them horizontally, Video 0 starts at 0_0 (top-left). Video 1 must start at the end of Video 0’s width, which is w0_0. Because Video 1 is shorter, you should use the fill option to paint the empty space black.

ffmpeg -i video1.mp4 -i video2.mp4 -filter_complex "[0:v][1:v]xstack=inputs=2:layout=0_0|w0_0:fill=black[v]" -map "[v]" output.mp4

Example: 2x2 Grid with Mixed Dimensions

Suppose you have four videos with different sizes: * Input 0: 1920x1080 (Top-left) * Input 1: 1280x720 (Top-right, positioned at X = 1920, Y = 0) * Input 2: 1280x720 (Bottom-left, positioned at X = 0, Y = 1080) * Input 3: 640x480 (Bottom-right, positioned at X = 1920, Y = 1080)

The layout string for this configuration is 0_0|w0_0|0_h0|w0_h0:

ffmpeg -i in1.mp4 -i in2.mp4 -i in3.mp4 -i in4.mp4 -filter_complex "[0:v][1:v][2:v][3:v]xstack=inputs=4:layout=0_0|w0_0|0_h0|w0_h0:fill=black[v]" -map "[v]" output.mp4

Method 2: Scaling Inputs First (Recommended)

Directly stacking mismatched resolutions can leave uneven black bars or empty gaps in your output grid. The cleanest way to stack videos of different dimensions is to scale them to uniform dimensions before passing them to the xstack filter.

The following command scales four different input videos to a uniform 640x360 resolution and stacks them in a perfect 2x2 grid:

ffmpeg -i in1.mp4 -i in2.mp4 -i in3.mp4 -i in4.mp4 -filter_complex \
"[0:v]scale=640:360[v0]; \
 [1:v]scale=640:360[v1]; \
 [2:v]scale=640:360[v2]; \
 [3:v]scale=640:360[v3]; \
 [v0][v1][v2][v3]xstack=inputs=4:layout=0_0|w0_0|0_h0|w0_h0[v]" \
-map "[v]" output.mp4

Method 3: Padding Inputs to Avoid Aspect Ratio Distortion

If scaling your videos stretches or distorts the image, you can use the pad filter. This adds black bars (pillarboxes or letterboxes) to each video to bring them to a uniform size before stacking.

This example pads two different videos to a uniform 1280x720 container, centering the original video, and stacks them side-by-side:

ffmpeg -i in1.mp4 -i in2.mp4 -filter_complex \
"[0:v]scale=1280:720:force_original_aspect_ratio=decrease,pad=1280:720:(ow-iw)/2:(oh-ih)/2[v0]; \
 [1:v]scale=1280:720:force_original_aspect_ratio=decrease,pad=1280:720:(ow-iw)/2:(oh-ih)/2[v1]; \
 [v0][v1]xstack=inputs=2:layout=0_0|w0_0[v]" \
-map "[v]" output.mp4