Skip to main content

Stream your Linux desktop to a Smart TV via Miracast/WFD, DLNA, or Chromecast

Project description

FluxCast

FluxCast streams a Linux desktop to a TV.

Release Python Platform Issues PRs

Demo

https://github.com/user-attachments/assets/ce01804c-2f86-4a5d-8ecf-d6f2a72f55d1

Project Status

Current validated scope:

  • wfd is the primary path and the only mode tested as release-ready.
  • dlna works as fallback.
  • cast is experimental and currently not working in the tested Samsung setup.

The project currently focuses on WFD/Miracast on Linux (Hyprland/wlroots class setups).
DLNA and Cast are available, but they are best treated as fallback or experimental paths.

Current limitation:

  • KDE/GNOME Wayland desktop capture uses xdg-desktop-portal in WFD mode.
  • For portal mode, install Python dependency dbus-next and allow screen-share in the desktop picker dialog.

Quick Start

Default WFD run (interactive monitor/peer selection):

python3 src/main.py

WFD with latency/session JSONL log:

python3 src/main.py --wfd-latency-log

DLNA fallback:

python3 src/main.py --protocol dlna --transport hls

Cast mode (optional, if your TV supports it):

python3 src/main.py --protocol cast

Force backend manually (if auto is not suitable on your session):

python3 src/main.py --capture-backend wf-recorder
python3 src/main.py --capture-backend x11grab
python3 src/main.py --protocol wfd --wfd-capture-backend portal
python3 src/main.py --protocol wfd --wfd-capture-backend wf-recorder
python3 src/main.py --protocol wfd --wfd-capture-backend x11grab

What Works Best

WFD (Primary)

screen + audio capture -> H.264/AAC RTP -> Wi-Fi Direct + RTSP -> TV WFD receiver

This is the lowest-latency and most predictable path in the current codebase.

DLNA (Fallback)

desktop capture -> HTTP stream -> DLNA/UPnP AVTransport -> native TV player
  • Prefer --transport hls on Samsung TVs.
  • progressive-ts can freeze or stutter on some models.

Cast (Optional)

  • Requires a TV/device with real Google Cast support.
  • Requires pychromecast.
  • Not reliable on many Samsung TV models.

Installation

AppImage

Download the latest FluxCast-x86_64.AppImage from the Releases page, then:

chmod +x FluxCast-x86_64.AppImage
./FluxCast-x86_64.AppImage

On first launch FluxCast will ask for your password once to install a system file for Wi-Fi Direct.

Depending on your desktop environment, you may need to install:

  • Hyprland / Sway: wf-recorder, ffmpeg
  • KDE / GNOME: gst-plugins-ugly (package name varies by distro)

PyPI

pip install fluxcast
sudo fluxcast-install-system

fluxcast-install-system installs the D-Bus policy, desktop entry, and system packages (GStreamer, ffmpeg, NetworkManager, etc.). Run it once after the pip install.

Arch Linux - AUR

yay -S fluxcast-git
# or
paru -S fluxcast-git

From source

git clone https://github.com/IlyaP358/fluxcast.git
cd fluxcast
pip install -r requirements.txt
sudo meta/install.sh
sudo systemctl reload dbus
sudo gtk-update-icon-cache /usr/share/icons/hicolor

[!WARNING] If PIP refuses to install the required libraries to your system, you will need to do that yourself using your distro's package manager.

DLNA/Cast features require additional packages listed in requirements.txt.

System tools (just as important)

WFD mode also depends on system binaries, not only Python packages:

  • ffmpeg
  • wf-recorder (Wayland/wlroots capture path)
  • xdg-desktop-portal (+ desktop backend: xdg-desktop-portal-kde / xdg-desktop-portal-gnome / xdg-desktop-portal-wlr)
  • nmcli, gdbus, iw, wpa_cli (Wi-Fi Direct and diagnostics)
  • pactl (audio monitor autodetect)

Use:

python3 src/main.py --doctor

to check your machine before running WFD.

Note: on KDE/GNOME Wayland, WFD auto backend now prefers portal first.

Note: on firewalld systems, FluxCast opens the WFD RTSP port (7236/tcp) for the duration of a session and closes it on exit (no-op without firewalld; disable with --wfd-no-firewall). See DOCUMENTATION.md -> "WFD and firewalld".

Documentation

Detailed flags, modes, and usage examples:
documentation/DOCUMENTATION.md

Tested Environment

Hardware:

ThinkBook 14 G4+ IAP
  • CPU: Intel i5-1240P (16 threads) up to 4.40 GHz
  • GPU: Intel Iris Xe Graphics
  • RAM: 16 GB
Dell XPS 13 Plus
  • CPU: Intel i5-1260P (16 threads) up to 4.70 GHz
  • GPU: Intel Iris Xe Graphics
  • RAM: 16 GB LPDDR5
ThinkPad T14 Gen 4
  • CPU: Intel i7-1355U (12 threads) up to 5.00 GHz
  • GPU: Intel Iris Xe Graphics
  • RAM: 32 GB

Software:

Arch Linux
  • Kernels: 7.0.8-arch1-1, 6.12.91-1-lts612
  • WMs: Hyprland (0.55.2)
  • DEs (for testing): KDE Plasma (6.6.5) | GNOME (50.1)
  • Shell: zsh (5.9), fish (4.7.1)
  • Terminal: kitty (0.46.2, 0.47.1)
CachyOS
  • Kernels: 7.0.3-1-cachyos
  • DEs (for testing): KDE Plasma (6.6.4)
  • Shell: bash (5.3.9)
  • Terminal: konsole (26.4.0)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fluxcast-0.1.5.tar.gz (349.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

fluxcast-0.1.5-py3-none-any.whl (180.4 kB view details)

Uploaded Python 3

File details

Details for the file fluxcast-0.1.5.tar.gz.

File metadata

  • Download URL: fluxcast-0.1.5.tar.gz
  • Upload date:
  • Size: 349.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.5

File hashes

Hashes for fluxcast-0.1.5.tar.gz
Algorithm Hash digest
SHA256 ae66a13f375d632f7f5114e93bcc321cf0de619f95e430777097a8c4a4b29219
MD5 898830ed6ba39554dbce1bd4912c4f7c
BLAKE2b-256 16280775ae372a5877c8205621830b64881fa464e9e45d2d58acad5b0c280b1c

See more details on using hashes here.

File details

Details for the file fluxcast-0.1.5-py3-none-any.whl.

File metadata

  • Download URL: fluxcast-0.1.5-py3-none-any.whl
  • Upload date:
  • Size: 180.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.5

File hashes

Hashes for fluxcast-0.1.5-py3-none-any.whl
Algorithm Hash digest
SHA256 1a1726388894812e01cefd1d49775f36cece0c94fa4e876c56660853b105bb85
MD5 bc0c9dc6e6378bd54ff7010f411d46b5
BLAKE2b-256 74ca44e953b43df4b42ef914664f8dbf3c5f637aeb46e6476655219442fd7dc5

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page