ffmpeg
Docker build for FFmpeg on Ubuntu / Alpine / Centos / Scratch / nvidia / vaapi
Top Related Projects
Mirror of https://git.ffmpeg.org/ffmpeg.git
The FFmpeg build script provides an easy way to build a static FFmpeg on OSX and Linux with non-free codecs included.
Port of FFmpeg with Emscripten
The fastest and safest AV1 encoder.
Quick Overview
The jrottenberg/ffmpeg repository is a Docker image project that provides a pre-built FFmpeg installation. It offers a convenient way to use FFmpeg in containerized environments, making it easier to deploy and use FFmpeg across different platforms without worrying about dependencies or system-specific installations.
Pros
- Easy deployment of FFmpeg in containerized environments
- Regularly updated with the latest FFmpeg versions
- Supports multiple architectures (x86_64, arm64, armhf)
- Includes various build options and additional libraries for extended functionality
Cons
- May have a larger image size compared to a minimal FFmpeg installation
- Requires Docker to be installed and running on the host system
- Limited customization options compared to building FFmpeg from source
- Potential security concerns when using pre-built images in sensitive environments
Getting Started
To use the jrottenberg/ffmpeg Docker image, follow these steps:
- Ensure Docker is installed on your system.
- Pull the latest image:
docker pull jrottenberg/ffmpeg - Run FFmpeg commands using the Docker container:
docker run -v $(pwd):/tmp/workdir jrottenberg/ffmpeg -i /tmp/workdir/input.mp4 /tmp/workdir/output.mp4
This example mounts the current directory to /tmp/workdir in the container and converts input.mp4 to output.mp4 using FFmpeg.
Competitor Comparisons
Mirror of https://git.ffmpeg.org/ffmpeg.git
Pros of FFmpeg
- Official source repository with the most up-to-date codebase
- Comprehensive documentation and extensive community support
- Wider range of features and customization options
Cons of FFmpeg
- Requires manual compilation and installation
- More complex setup process for beginners
- Larger repository size due to full source code and development history
Code Comparison
FFmpeg:
int main(int argc, char **argv) {
AVFormatContext *fmt_ctx = NULL;
AVCodecContext *dec_ctx = NULL;
AVStream *video_stream = NULL;
const char *src_filename = NULL;
jrottenberg/ffmpeg:
FROM alpine:3.17 as base
RUN apk add --no-cache --update libgcc libstdc++ ca-certificates libcrypto1.1 libssl1.1 libgomp expat
FROM base AS build
The FFmpeg repository contains the actual source code of the FFmpeg project, while jrottenberg/ffmpeg is a Docker image that packages FFmpeg for easy deployment. The code snippets show this difference, with FFmpeg displaying C code for video processing and jrottenberg/ffmpeg showing a Dockerfile for building the container image.
jrottenberg/ffmpeg simplifies the deployment process, especially for containerized environments, but may have limitations in terms of customization and access to the latest features compared to the official FFmpeg repository.
Pros of FFmpeg-Builds
- Offers a wider range of build options and configurations
- Provides more up-to-date builds with the latest FFmpeg features
- Includes builds for multiple platforms and architectures
Cons of FFmpeg-Builds
- Less focus on Docker integration compared to ffmpeg
- May have a steeper learning curve for users new to FFmpeg
- Potentially larger file sizes due to more comprehensive builds
Code Comparison
FFmpeg-Builds:
./build.sh --target=win64-gpl --enable-gpl --enable-version3 --enable-sdl2
ffmpeg:
FROM jrottenberg/ffmpeg:4.3-alpine
RUN ffmpeg -version
The FFmpeg-Builds example shows a build script with custom options, while the ffmpeg example demonstrates a simple Docker usage.
The FFmpeg build script provides an easy way to build a static FFmpeg on OSX and Linux with non-free codecs included.
Pros of ffmpeg-build-script
- Offers more customization options for building FFmpeg
- Supports building on a wider range of operating systems
- Provides a more up-to-date FFmpeg build with the latest features
Cons of ffmpeg-build-script
- Requires more manual setup and configuration
- May have longer build times due to compiling from source
- Less suitable for quick deployments or containerized environments
Code Comparison
ffmpeg-build-script:
./build-ffmpeg --build
ffmpeg:
FROM jrottenberg/ffmpeg:4.4-ubuntu
Summary
ffmpeg-build-script is better suited for users who need custom FFmpeg builds with specific features or optimizations. It offers more flexibility but requires more setup time and technical knowledge.
ffmpeg is ideal for users who need a pre-built FFmpeg solution, especially in containerized environments. It's quicker to deploy but offers less customization.
The choice between the two depends on the specific use case, with ffmpeg-build-script favoring customization and ffmpeg prioritizing ease of use and quick deployment.
Port of FFmpeg with Emscripten
Pros of ffmpeg.js
- Runs in the browser, enabling client-side video processing without server-side dependencies
- Lightweight and portable, suitable for web applications and environments where installing FFmpeg is not feasible
- Provides a JavaScript API for easier integration with web applications
Cons of ffmpeg.js
- Limited functionality compared to the full FFmpeg suite
- May have performance limitations for complex video processing tasks in the browser
- Requires additional setup and configuration for use in web projects
Code Comparison
ffmpeg.js:
const ffmpeg = require('ffmpeg.js');
const result = ffmpeg({
MEMFS: [{ name: "test.webm", data: testData }],
arguments: ["-i", "test.webm", "-c:v", "libx264", "out.mp4"],
});
ffmpeg:
ffmpeg -i input.webm -c:v libx264 output.mp4
Summary
ffmpeg.js is a JavaScript port of FFmpeg that runs in the browser, offering client-side video processing capabilities. It's lightweight and portable but has limited functionality compared to the full FFmpeg suite. The ffmpeg repository provides the complete FFmpeg toolset, offering more extensive features and better performance for complex tasks, but requires server-side installation and execution.
The fastest and safest AV1 encoder.
Pros of rav1e
- Specialized AV1 encoder, potentially offering better AV1-specific optimizations
- Written in Rust, providing memory safety and performance benefits
- More focused development on AV1 encoding, potentially leading to faster updates for this specific codec
Cons of rav1e
- Limited to AV1 encoding, lacking the versatility of FFmpeg's multi-codec support
- Smaller community and ecosystem compared to FFmpeg's extensive user base and plugins
- May require additional tools for full video processing workflows, unlike FFmpeg's all-in-one solution
Code Comparison
rav1e (Rust):
use rav1e::prelude::*;
let cfg = Config::default();
let mut ctx: Context<u8> = cfg.new_context().unwrap();
let frame = ctx.new_frame();
ctx.send_frame(frame).unwrap();
FFmpeg (C):
AVCodecContext *c = avcodec_alloc_context3(codec);
AVFrame *frame = av_frame_alloc();
avcodec_open2(c, codec, NULL);
av_init_packet(&pkt);
avcodec_send_frame(c, frame);
Both repositories serve different purposes within the video encoding ecosystem. rav1e focuses exclusively on AV1 encoding, leveraging Rust's benefits, while FFmpeg offers a comprehensive suite of audio and video processing tools with support for multiple codecs. The choice between them depends on specific project requirements and the desired level of specialization versus versatility.
Convert 
designs to code with AI
Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.
Try Visual CopilotREADME
FFmpeg Docker image
This project prepares a minimalist Docker image with FFmpeg. It compiles FFmpeg from sources following instructions from the Compilation Guide.
You can install the latest build of this image by running docker pull jrottenberg/ffmpeg:${VERSION}-${VARIANT} or docker pull ghcr.io/jrottenberg/ffmpeg:${VERSION}-${VARIANT}.
This image can be used as a base for an encoding farm.
Builds / Avaliabvle Docker Containers
There are different builds available: Below is a table that provides examples for the nomenclature:
ffmpeg-<version>-<os variant and version>
| image name | OS ver | ffmpeg ver | variant | description |
|---|---|---|---|---|
| ffmpeg-7.1-ubuntu2404 | 24.04 | 6.x - 7.x | ubuntu | external libraries are installed from os packages, and ffmpeg is built from source. See Ubunu Compilation Guide for details on this. |
| ffmpeg-7.1-ubuntu2404-edge | 24.04 | 6.x - 7.x | ubuntu | libs and ffmpeg are built from source. See Ubunu Compilation Guide for details on this. |
| ffmpeg-7.1-vaapi2404 | 24.04 | 6.x - 7.x | ubuntu | like: ubuntu2404 but enables: Video Acceleration API (VAAPI) in ffmpeg |
| ffmpeg-7.1-nvidia2204-edge | 22.04 | 6.x - 7.x | ubuntu | Built w/ NVIDIA's hardware-accelerated encoding and decoding APIs enabled |
| ffmpeg-7.1-alpine320 | 3.20 | 6.x - 7.x | alpine | vendor libs, but ffmpeg is built from source |
| ffmpeg-7.1-scratch | 3.20 | 6.x - 7.x | alpine | vendor libs, and ffmpeg are built from source. Also we make the distro as small as possible by not installing any packages in base and striping symbols of installed libs |
ffmpeg <version> can be one of the following: 6.1, 7.0, 7.1 with the above table.
Note: The current versions of ffmpeg supported are anything newer than 3 years old and not exceeded the end-of-life
Here are some additional older builds
- alpine based images
ffmpeg:<version>-alpineorffmpeg:<version>-alpine313(old versions withffmpeg:<version>-alpine312,ffmpeg:<version>-alpine311)- alpine based scratch images
ffmpeg:<version>-scratchorffmpeg:<version>-scratch313(old versions withffmpeg:<version>-scratch312,ffmpeg:<version>-scratch311)
- alpine based scratch images
- ubuntu based images
ffmpeg:<version>-ubuntuorffmpeg:<version>-ubuntu2004(old versions withffmpeg:<version>-ubuntu1804,ffmpeg:<version>-ubuntu1604)- ubuntu based nvidia images
ffmpeg:<version>-nvidiaorffmpeg:<version>-nvidia2004(old versions withffmpeg:<version>-nvidia1804,ffmpeg:<version>-nvidia1604) - ubuntu based vaapi images
ffmpeg:<version>-vaapi1804orffmpeg:<version>-vaapi2004(old versions withffmpeg:<version>-vaapi1804,ffmpeg:<version>-nvidia1604)
- ubuntu based nvidia images
Philosophy behind the different builds
ubuntu2404 We chose Ubuntu 24.04 because it is the LTS ( Long Term Support ) build of Ubuntu. We used the ffmpeg support libraries from the Ubuntu distrobution where possible. Example: we use 'libx264-dev' as the package to install. We do not tie it to a version. This way when its time to update from 24.04 to 26.04 we can simply update the base Docker template for ubuntu. This will make updating the OS easier as time goes on.
ubuntu2404-edge
This image is just like the above ubuntu2404 container image, except we build all of the ffmpeg support libraries. This is in the spirit of the original intent of this project jrottenberg/ffmpeg alltogether. Building everything that ffmpeg needs, and ffmpeg itself from source. This gives us the most control over all of the details of release. The drawback of this is that its much harder to keep updated. The thought process of having both 'Ubuntu-2404' and 'Ubuntu-2404-edge' is that it makes updating the OS easier over time.
vaapi2404
This release is like also ubuntu2404 but enables: Video Acceleration API (VAAPI) when building ffmpeg
nvidia2204-edge
This release is like also ubuntu2404 but enables: NVIDIA's hardware-accelerated encoding and decoding APIs enabled
alpine320 alpine uses the os vendor libs, but ffmpeg is built from source.
scratch Scratch is also an alpine image. We build the vendor libs, and ffmpeg from source. Also we make the distro as small as possible by not installing any packages in base and striping symbols of installed libs.
Generate list of recent Docker Container Images
You can use the following command to generate a list of current images:
$ python3 -mvenv .venv
$ source .venv/bin/activate
$ pip install requests
$ python3 ./generate-list-of-recent-images.py > list_of_recent_images.txt
$ deactivate
$ rm -rf .venv
$ less list_of_recent_images.txt
If you want to compare the one you have locally, use the following command:
$ docker images | grep ffmpeg | sort | awk '{print $1 ":" $2 "\t" $7 $8}'
Please use Github issues to report any bug or missing feature.
Test
ffmpeg version N-98740-ga72d529 Copyright (c) 2000-2020 the FFmpeg developers
built with gcc 7 (Ubuntu 7.5.0-3ubuntu1~18.04)
configuration: --disable-debug --disable-doc --disable-ffplay --enable-shared --enable-avresample --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-gpl --enable-libass --enable-fontconfig --enable-libfreetype --enable-libvidstab --enable-libmp3lame --enable-libopus --enable-libtheora --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libxcb --enable-libx265 --enable-libxvid --enable-libx264 --enable-nonfree --enable-openssl --enable-libfdk_aac --enable-postproc --enable-small --enable-version3 --enable-libbluray --enable-libzmq --extra-libs=-ldl --prefix=/opt/ffmpeg --enable-libopenjpeg --enable-libkvazaar --enable-libaom --extra-libs=-lpthread --enable-libsrt --enable-libaribb24 --enable-vaapi --extra-cflags=-I/opt/ffmpeg/include --extra-ldflags=-L/opt/ffmpeg/lib
libavutil 56. 58.100 / 56. 58.100
libavcodec 58.100.100 / 58.100.100
libavformat 58. 51.100 / 58. 51.100
libavdevice 58. 11.101 / 58. 11.101
libavfilter 7. 87.100 / 7. 87.100
libavresample 4. 0. 0 / 4. 0. 0
libswscale 5. 8.100 / 5. 8.100
libswresample 3. 8.100 / 3. 8.100
libpostproc 55. 8.100 / 55. 8.100
configuration:
--disable-debug
--disable-doc
--disable-ffplay
--enable-shared
--enable-avresample
--enable-libopencore-amrnb
--enable-libopencore-amrwb
--enable-gpl
--enable-libass
--enable-fontconfig
--enable-libfreetype
--enable-libvidstab
--enable-libmp3lame
--enable-libopus
--enable-libtheora
--enable-libvorbis
--enable-libvpx
--enable-libwebp
--enable-libxcb
--enable-libx265
--enable-libxvid
--enable-libx264
--enable-nonfree
--enable-openssl
--enable-libfdk_aac
--enable-postproc
--enable-small
--enable-version3
--enable-libbluray
--enable-libzmq
--extra-libs=-ldl
--prefix=/opt/ffmpeg
--enable-libopenjpeg
--enable-libkvazaar
--enable-libaom
--extra-libs=-lpthread
--enable-libsrt
--enable-libaribb24
--enable-vaapi
--extra-cflags=-I/opt/ffmpeg/include
--extra-ldflags=-L/opt/ffmpeg/lib
Capture output from the container to the host running the command
docker run jrottenberg/ffmpeg:4.4-alpine \
-i http://url/to/media.mp4 \
-stats \
$ffmpeg_options - > out.mp4
Examples
Extract 5s @00:49:42 into a GIF
docker run jrottenberg/ffmpeg:4.4-alpine -stats \
-i http://archive.org/download/thethreeagesbusterkeaton/Buster.Keaton.The.Three.Ages.ogv \
-loop 0 \
-final_delay 500 -c:v gif -f gif -ss 00:49:42 -t 5 - > trow_ball.gif
Convert 10bits MKV into a 10Bits MP4
docker run -v $(pwd):$(pwd) -w $(pwd) jrottenberg/ffmpeg:4.4-scratch \
-stats \
-i http://www.jell.yfish.us/media/jellyfish-20-mbps-hd-hevc-10bit.mkv \
-c:v libx265 -pix_fmt yuv420p10 \
-t 5 -f mp4 test.mp4
The image has been compiled with X265 Multilib. Use the pixel format switch to change the number of bits per pixel by suffixing it with 10 for 10bits or 12 for 12bits.
Convert a local GIF into a mp4
Let's assume original.gif is located in the current directory :
docker run -v $(pwd):$(pwd) -w $(pwd)\
jrottenberg/ffmpeg:4.4-scratch -stats \
-i original.gif \
original-converted.mp4
Use ZeroMQ to toggle filter value on-fly
Let's start some process continuously writing some radio music, and listen it:
docker run --rm -d -v $(pwd):$(pwd) -w $(pwd) -p 11235:11235 \
--name radio-writer jrottenberg/ffmpeg:4.4-alpine \
-i http://radio.casse-tete.solutions/salut-radio-64.mp3 \
-filter_complex '[0:a]volume@vol=1,azmq=bind_address=tcp\\\://0.0.0.0\\\:11235[out]' \
-map '[out]' ./salut-radio.mp3
docker run -it -v $(pwd):$(pwd) -w $(pwd) --entrypoint=ffprobe jrottenberg/ffmpeg:4.4-alpine -v quiet -show_streams salut-radio.mp3
Now, just toggle its volume on-fly, and hear how it changes:
docker run --rm --network=host --entrypoint sh jrottenberg/ffmpeg:4.4-ubuntu -c \
'echo "volume@vol volume 2" | zmqsend -b tcp://127.0.0.1:11235'
Send a stream over SRT
Let's send video.mp4 to srt-listener on port 9000 over SRT protocol.
docker run -v $(pwd):$(pwd) jrottenberg/ffmpeg:4.4-centos \
-re -i $(pwd)/video.mp4 -acodec copy -vcodec copy -f mpegts srt://srt-listener:9000?pkt_size=1316
Use hardware acceleration enabled build
Thanks to qmfrederik for the vaapi ubuntu based variant
jrottenberg/ffmpeg:vaapi or jrottenberg/ffmpeg:${VERSION}-vaapi
- Run the container with the device attached /dev/dri from your host into the container :
docker run --device /dev/dri:/dev/dri -v $(pwd):$(pwd) -w $(pwd) jrottenberg/ffmpeg:4.4-vaapi [...]
- Have the Intel drivers up and running on your host. You can run
vainfo(part of vainfo package on Ubuntu) to determine whether your graphics card has been recognized correctly. - Run ffmpeg with the correct parameters, this is the same as when running ffmpeg natively.
Use nvidia hardware acceleration enabled build
Thanks to ShaulMyplay for the nvidia based variant
Supports nvenc only on all ffmpeg versions, and hardware decoding and scaling on ffmpeg >= 4.0
- Install nvidia latest drivers on host machine.
- Install nvidia-docker on host machine.
- Run container using "--runtime=nvidia" flag and use supported ffmpeg hwaccel options
Hardware encoding only example:
docker run --runtime=nvidia jrottenberg/ffmpeg:4.4-nvidia -i INPUT -c:v nvenc_h264 -preset hq OUTPUT
Full hardware acceleration example:
docker run --runtime=nvidia jrottenberg/ffmpeg:4.4-nvidia -hwaccel cuvid -c:v h264_cuvid -i INPUT -vf scale_npp=-1:720 -c:v h264_nvenc -preset slow OUTPUT
See what's inside the beast
docker run -it --entrypoint='bash' jrottenberg/ffmpeg:7.1-ubuntu2404
for i in ogg amr vorbis theora mp3lame opus vpx xvid fdk x264 x265;do echo $i; find /usr/local/ -name *$i*;done
Libs are in /lib in alpine.
docker run -it --entrypoint='sh' jrottenberg/ffmpeg:7.1-alpine320
for i in ogg amr vorbis theora mp3lame opus vpx xvid fdk x264 x265;do echo $i; find /lib/ -name *$i*;done
FFMPEG Supported Libraries
The following libraries are used by FFMPEG. The version number and release date are provided along with the license information. These version numbers are for the lib source builds, which are 'ubuntu2404-edge' and 'foo'. These libs are included in the package images as well, but the version numbers might vary slightly.
| Libraries | Version | Release Date | License |
|---|---|---|---|
| ffmpeg | 7.1 | GNU Lesser General Public License (LGPL) version 2.1 | |
| libogg | 1.3.4 | 08-2019 | BSD-style license |
| libopencore-amr | 0.1.6 | 08-2022 | Apache License |
See generate-source-of-truth-ffmpeg-versions.py to update a version
FFMPEG Supported Libraries
The following libraries are used by FFMPEG. The version number and release date are provided along with the license information. These version numbers are for the lib source builds, which are 'ubuntu2404-edge' and 'foo'. These libs are included in the package images as well, but the version numbers might vary slightly.
Contribute
Legal
Those docker images use code of <a href=http://ffmpeg.org>FFmpeg licensed under the <a href=http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html>LGPLv2.1 and their source can be downloaded on <a href=https://github.com/jrottenberg/ffmpeg>github.com/jrottenberg/ffmpeg.
Top Related Projects
Mirror of https://git.ffmpeg.org/ffmpeg.git
The FFmpeg build script provides an easy way to build a static FFmpeg on OSX and Linux with non-free codecs included.
Port of FFmpeg with Emscripten
The fastest and safest AV1 encoder.
Convert designs to code with AI
Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.
Try Visual Copilot