Install graphviz with Homebrew, apk, chocolatey, apt, dnf, MacPorts, Nix, pacman, scoop, winget, zypper

Project links

• Homepage
• Repository
• Upstream docs
• Package manager page

Commands and aliases

• acyclic
• bcomps
• ccomps
• circo
• cluster
• dijkstra
• dot
• dot2gxl
• dot_builtins
• dot_sandbox
• edgepaint
• fdp
• gml2gv
• graphml2gv
• gv2gml
• gv2gxl
• gvcolor
• gvgen
• gvmap
• gvmap.sh
• gvpack
• gvpr
• gxl2dot
• gxl2gv
• mm2gv
• neato
• nop
• osage
• patchwork
• prune
• sccmap
• sfdp

Project history

The project's roots go back to research on drawing directed graphs for software engineering and structured data. The Graphviz history discussion by project participants describes the initial kernel as an attempt to draw software-engineering graphs, with ranking via network simplex becoming central to directed graph layout.

AT&T Bell Laboratories research papers from the early 1990s formalized the dot approach. The 1993 directed-graph drawing paper describes a four-pass algorithm for rank assignment, crossing reduction, coordinate assignment, and spline edge routing, and the later Graphviz toolkit paper frames the project as a set of stream-oriented filters, libraries, and interactive front ends.

By the early 2000s, Graphviz had expanded from dot into a family of engines and tools: neato, twopi, circo, fdp, sfdp, gvpr, sccmap, unflatten, and others. The official credits page records the long-running roles of Emden Gansner, Stephen North, John Ellson, Yifan Hu, Eleftherios Koutsofios, Gordon Woodhull, and later maintainers.

Adoption history

Graphviz spread because it fit Unix package culture: users could write a small DOT file, pipe it through a renderer, and get PostScript, SVG, PDF, PNG, image maps, or other output without a heavyweight diagram editor. Its stream model made it easy for other tools to emit DOT as an interchange format.

The official download page documents broad operating-system packaging, including Debian and Ubuntu apt, Fedora and related dnf packages, Windows installers, and GitLab-hosted release artifacts. Homebrew, MacPorts, Nix, Scoop, Chocolatey, distro packages, and language-specific wrappers made Graphviz a common transitive dependency in developer environments.

How it is used

Typical CLI usage is to save a graph in DOT syntax and run an engine such as dot, neato, fdp, sfdp, circo, or twopi to produce a rendered artifact. Supporting commands such as tred, unflatten, sccmap, gvpack, gvpr, and ccomps make Graphviz useful as a graph-processing toolkit, not only as a renderer.

Graphviz is often installed so other tools can find the dot executable. Python, R, JavaScript, documentation, modeling, reverse-engineering, and CI workflows commonly generate DOT and delegate layout to the packaged Graphviz binary or libraries.

Why package nerds care

Graphviz is a classic example of a small CLI surface hiding a large native dependency stack: C libraries, plugins, font and image handling, multiple layout engines, and platform-specific packaging decisions. Its formulae and distro packages are watched because many higher-level packages fail in surprising ways when dot is missing.

The package is also a useful test case for long-lived open-source infrastructure moving from research lab origins into modern forge hosting. The official GitLab repository records a project imported into GitLab in 2017, while the documented algorithms and manpages preserve a much older Bell Labs lineage.

Timeline

• 1988: The DAG directed-graph drawing program appears in the research lineage cited by Graphviz papers.
• 1993: The directed graph drawing algorithm behind dot is published in IEEE Transactions on Software Engineering.
• 1999: The Graphviz toolkit paper describes libraries, layout tools, graph filters, and stream-oriented design.
• 2003: Graphviz and Dynagraph are presented as static and dynamic graph drawing tools from AT&T Labs Research.
• 2017: The official Graphviz source repository is created on GitLab.
• 2024: The official credits page still records the multi-decade contributor base behind dot, neato, plugins, build work, and related tools.

Related projects

• DOT language: the textual graph description language consumed by Graphviz layout engines.
• Dynagraph: a related AT&T Labs project for dynamic and incremental graph layout.
• graphviz2drawio, pydot, PyGraphviz, d3-graphviz, and many language bindings depend on Graphviz concepts, binaries, or libraries.

Sources

• https://forum.graphviz.org/t/the-history-of-graphviz/765
• https://gitlab.com/graphviz/graphviz
• https://graphviz.org/credits/
• https://graphviz.org/documentation/
• https://graphviz.org/documentation/EGKNW03.pdf
• https://graphviz.org/documentation/GN99.pdf
• https://graphviz.org/documentation/TSE93.pdf
• https://graphviz.org/download/