Country
Where do you live?
European responses climbed to 60.0%, up from 52% a year earlier, while North America slipped from 27.5% to 21.3%. The remaining regions each moved by less than a point.
NixOS Community · Annual Report
2025 Community Survey Results
3,399 responses · September 1 — December 15, 2025
The 2025 NixOS Community Survey ran from September 1 to December 15, 2025, drawing 3,399 full responses — a 48% increase over the 2,290 received in 2024.
The growth reflects a noticeably newer audience: 28.9% of respondents started using Nix in the past year, and 82.5% have under four years of experience. New questions in this year's survey, and the additional charts that accompany them, were added to surface insights into who that new majority is, what tools they use, and where they look for help.
People
Categories with fewer than 5 respondents are not shown.
Age
What is your age in years?
The age curve flattened. The 25-34 cohort, still the largest, fell from 35.6% to 30.7%, and 35-44 dropped from 22.7% to 19.4%. Every group under 25 or over 44 grew, by between 0.4 and 3.1 points — a modest but broad-based widening.
Gender Identity
What is your gender identity?
The community is overwhelmingly male: 78.3% of respondents identify as men. Women account for 7.2% and non-binary respondents 5.9%; a further 6.6% declined to answer.
Transgender Identity
Do you identify as transgender?
A majority (81.8%) do not identify as transgender; 9.8% do, and 7.2% declined to answer.
Years Coding
How many years of programming experience do you have (including time spent learning or studying)?
Reported programming experience leans long: most respondents fall in the 1–14 year bands. Groups with four or fewer years of experience grew compared with the prior year, while groups with five or more years shrank. This year's instrument also splits the 10+ bracket into finer bins, surfacing more detail at the experienced end.
Role
Which of the following best describes your primary role or occupation?
Students lead at 24.0%, followed by full-stack (13.4%) and back-end developers (10.7%); system and network administrators trail at 4.8%. A further 10.3% selected "Other".
Industry
What best describes your economic sector?
"Other" is the largest single category at 21.0%, suggesting the listed industries miss common workplaces. Academia or education (17.6%) is the largest named category. Telecom (5.9%), Consulting (5.8%), and Financial services (4.7%) follow.
Domain
What best describes your software domain?
"Other" leads at 15.7%, suggesting the listed domains miss common categories. Web development (13.8%), open-source software development (10.8%), and programming languages / compilers / code analysis (9.2%) follow.
Technology
Operating Systems
Which operating systems do you regularly use on your devices (e.g. for development, work, or personal use)?
Nearly all respondents — 96.9% — use Linux on at least one device. Android (57.8%) and Windows (35.5%) are next, with macOS at 24.4% and iOS at 19.6%. Categories below 0.5% or with fewer than 5 respondents are combined into the "Other (combined)" bar.
Operating systems running Nix
On which operating systems do you currently use Nix?
Among respondents who use Nix, NixOS itself is the primary host (90.0%). Other GNU/Linux distributions (26.5%), macOS (18.1%), and Windows via WSL (11.0%) round out the meaningful platforms. 3.7% report not using Nix at all.
Target Triple
Which target triple best describes your system? Run "uname -a" if you are not sure.
x86_64-linux-gnu: 94.5%. ARM combined (aarch64-linux-gnu 17.4% +
aarch64-apple-darwin 16.2%) reaches 33.6%. RISC-V: 2.0%.
Experimental Features
Which Nix experimental features do you use?
Flakes (78.9%) and the new nix CLI (66.7%) lead. The next-ranked feature
(pipe operators, 8.0%) trails by 58.7 points. 10.9% report using no
experimental features; 7.9% are not sure.
Install Method
How have you installed Nix on any of your systems?
Most respondents (78.4%) first met Nix through a NixOS installation.
The official install script accounts for another 36.0%, followed by
the DeterminateSystems/nix-installer at 15.8%. Self-built and
third-party-packaged installations together cover the remaining 13.6%.
Nix Implementations
Which of these do you use?
Upstream Nix remains the default by a wide margin (87.5%). Lix (14.8%) and Determinate Nix (9.8%) are the most-used alternatives. Tvix (0.8%) and Snix (1.2%) are still in single digits.
Infrastructure
On what types of systems or infrastructure do you use Nix?
Personal computers (90.3%) are the dominant target. Workstations (55.8%) and home servers (52.6%) follow. Production servers (21.4%) and CI environments (20.5%) are used at roughly a quarter the rate of personal machines.
Nix Version
Which version of Nix do you use most often? (Enter the full SemVer version, e.g. 2.18.5. You can check by running `nix-env --version`.)
The 2.28 series dominates the answers that could be parsed: 2.28.5 alone accounts for 21.7%, with 2.28.4 at 10.5%. Roughly a third of respondents skipped the question; a further 6.4% gave an unparseable answer.
Versions are extracted with a SemVer regex against the free-text response. Unparseable answers are labelled "No Match" and bins below 0.1% or 5 respondents are folded into "Other (combined)".
NixOS Release
Which NixOS releases do you currently use on any of your systems?
25.05 (the current stable) leads at 61.6%, followed closely by unstable at 56.8%. The two combined exceed 100%, indicating many respondents run more than one release across their machines. Older releases drop off sharply: 24.11 at 6.6%, 24.05 at 2.2%, and 23.11 or earlier at 1.4%.
Hardware Configuration
How do you configure hardware on NixOS?
nixos-generate-config is still the dominant entry point at 68.6%,
with manual configuration cited by 39.0% and the nixos-hardware
repository by 26.7%. nixos-facter lands at 3.9%.
Software Ecosystems
Which software ecosystems do you use Nix with?
NixOS configurations leads at 81.1% (a meta-result — Nix users using Nix for their own systems). Among programming languages: Python 45.8%, Rust 42.9%, Bash 36.8%, C 26.7%, JavaScript 23.5%, C++ 22.7%, Go 21.3%. Categories below 0.5% or with fewer than 5 respondents are combined into the "Other (combined)" bar.
Experience
Stable Upgrade
If you use NixOS stable, how was your experience upgrading to the new stable version (25.05)?
Among respondents who answered, most had a clean upgrade: 45.4% report no issues and 23.7% only minor ones, while moderate, severe-resolved, and severe-unresolved issues together account for 8.6%. A further 15.5% had not yet upgraded at survey time, and 6.7% were unaware a new stable release existed.
The Sankey funnel below re-stages these responses as a flow — from all respondents, to whether they knew a new stable release existed, to whether they upgraded, to the severity they hit. This staging is a derived analyst interpretation, not a survey field: the single question cannot truly separate "knew but chose not to upgrade" from other reasons for staying put, so the funnel is exploratory and meant to be read visually rather than as exact sub-populations. The companion Sankey links experience (years using Nix) to the upgrade outcome.
- I had no issues.
- I had minor issues.
- I had moderate issues.
- I had severe issues but figured it out after some time.
- I had severe issues and could not make the upgrade.
- I did not know there was a new stable release.
- I have not upgraded.
Involvement
Which of these best describe your involvement with the Nix ecosystem?
Nearly every respondent uses NixOS (87.1%), and two-thirds (66.7%) install software with Nix more broadly. Roughly half maintain NixOS machines (58.1%) or develop software on Nix (48.8%). Direct contribution to Nixpkgs is much narrower: 22.8% have submitted packages or patches, 15.5% maintain packages, and 2.9% hold merge access.
Years using Nix
How many years have you been using Nix (at any level of involvement)?
The community skews new: 28.9% have used Nix for less than a year, 34.0% for one to two years, and 19.6% for three to four. Combined, 82.5% of respondents have under four years of Nix experience. Only 12.3% have used Nix for five or more years.
Nix Skill Level
How would you describe your skill level with Nix?
Self-described skill remains roughly stable year over year: Intermediate (45.6%) is the largest group, followed by Beginner (38.9%) and Advanced (10.2%).
First heard of: Nix or NixOS
Did you hear about Nix or NixOS first? If you heard about both at the same time, select which one first made you interested in the Nix ecosystem.
NixOS is the more common entry point into the ecosystem: 70.8% of respondents heard about it first, against 20.0% who heard of Nix first. 7.6% no longer remember.
First-exposure channel
How did you first hear about Nix or NixOS?
YouTube is the single largest channel (28.3%), well ahead of word-of-mouth from a friend or colleague (15.9%). 12.0% no longer remember. Blogs (7.3%), search (6.7%), Reddit (4.9%), and work introductions (4.6%) make up the next tier. Categories below 0.5% or fewer than 5 respondents are folded into "Other (combined)".
Discovery flow
Did you hear about Nix or NixOS first? If you heard about both at the same time, select which one first made you interested in the Nix ecosystem. → How did you first hear about Nix or NixOS?
This alluvial view links whether respondents first heard of Nix or NixOS (left) to the channel that introduced them (right). It re-expresses the two discovery questions as a single flow: NixOS-first respondents dominate every channel, and YouTube and word-of-mouth carry the heaviest streams. Channels reaching fewer than 5 respondents in a given flow are dropped, so only the substantial paths are drawn.
User types
Which user types do you identify with?
"Love the idea" (A) is by far the most-claimed identity, picked by 85.3% of respondents. Roughly half (51.2%) also identify as "get things done" pragmatists (C), and 44.2% as curious learners (B). Aspiring or active contributors (D) account for 31.3%, and strategic decision-makers (E) for 14.8%. Because the question allows multiple selections, respondents often combine personas — e.g., enthusiast and pragmatist. The UpSet chart below the bars makes these combinations explicit: each column is one exact set of personas, the connected dots show which personas it contains, and the bar above counts how many respondents chose exactly that combination. Combinations with fewer than 5 respondents are not shown.
Persona definitions (from the survey prompt):
A. Love the idea behind Nix or NixOS. Maybe you spread the word among friends and coworkers. Interested in or enthusiastic about free and open source software, Linux, home automation, online communities, or distro-hopping.
B. Curious about Nix and how it works. Maybe you enjoy or want to learn functional programming, or just want to learn new things. Identify as: Nix-curious developer, student of a technical field, educator, or academic researcher.
C. Use Nix or NixOS to get things done or boost your team's productivity, learn it to grow career opportunities, or use it on the job because someone said so. Identify as: system administrator, employee developer, DevOps engineer, natural scientist, open source software author, or early-career professional.
D. Work or want to work on the Nix ecosystem rather than just with it. At least one of: aspiring contributor, novice contributor, drive-by contributor, package maintainer, code owner, community team member, or sponsor.
E. Make strategic decisions for your team or company: which technologies to adopt, which skills to train, which projects to support or invest in, which services or products to offer. Examples: entrepreneur, team lead, software architect, CTO, sales executive, public service administrator.
UpSet row key (short code → full statement):
- A — A. I love the idea behind Nix
- B — B. I'm curious how Nix works
- C — C. I use Nix to get things done
- D — D. I (want to) work on Nix
- E — E. I'm a decision-maker
Self-described skills
Which of these describe you?
Respondents pick the traits that describe their relationship with Nix across a 12-item checklist. Tutorial reliance leads at 62.5%, followed by using the module system for one's own configuration (56.3%) and having used overlays (53.7%). The share that feels comfortable contributing upstream (23.3%), reads the Nix source (19.4%), or claims to understand every error message (4.0%) drops sharply.
Wished-for improvements
You can afford about 3 improvements. Which aspects of the Nix package manager do you pick?
Asked to pick about three areas they would improve, respondents prioritise flakes (50.3%) — the most-cited single area — alongside error messages (47.6%) and the reference manual (41.5%). Performance (39.2%) and language usability (31.2%) form the next tier. Platform expansion (Windows 6.4%, FreeBSD 3.5%) ranks low.
Regular toolset
Do you consider Nix part of your regular toolset?
A clear majority — 81.1% — consider Nix part of their regular toolset. 15.2% do not, and 3.7% skipped the question.
Help-search success rate
When you look for help with Nix or NixOS, how often do you find a useful or acceptable answer?
Respondents largely find what they need: 48.8% say they reach a useful answer "often" and another 6.7% "always". 30.4% land on "sometimes", while only a small minority report rare success (4.5%) or none at all (0.3%).
Involvement rate by years using Nix
Which of these best describe your involvement with the Nix ecosystem? × How many years have you been using Nix (at any level of involvement)?
Among respondents in each experience band, the share reporting each involvement role. Bands exclude non-users and undisclosed experience.
Involvement lift by years using Nix
Which of these best describe your involvement with the Nix ecosystem? × How many years have you been using Nix (at any level of involvement)?
How much more (or less) likely each involvement role is within an experience band relative to the overall baseline (1.0× = no difference). Bands exclude non-users and undisclosed experience.
Expertise vs Experience
How would you describe your skill level with Nix? × How many years have you been using Nix (at any level of involvement)?
Each row sums to 100%. Beginners are concentrated in <2 years (88.0%); Intermediate peaks at 1-2 years (43.3%) and extends through 3-4 years (28.0%); Advanced spreads from 3 years onward, peaking at 3-4 years (30.3%). Self-described skill and time-spent aren't strictly aligned — 2.0% of Advanced respondents report <1 year using Nix and 2.3% of Beginners report 5+ years.
Experience vs Expertise
How many years have you been using Nix (at any level of involvement)? × How would you describe your skill level with Nix?
The complement of the previous chart: each row (a tenure cohort) sums to 100%, showing how self-described skill distributes within that cohort. Useful for the dual question — "what does this experience level look like in skill terms?" rather than "what experience does this skill level have?"
Trait rate by skill level
Which of these describe you? × How would you describe your skill level with Nix?
Each cell: of respondents at this skill level, what percent picked this trait. Cells in a column do NOT sum to 100% — traits are multi-choice.
Beginners overwhelmingly rely on tutorials (83%); rate drops to 57% at Intermediate and 18% at Advanced. Writing own NixOS modules, understanding overlays, and contributing upstream all climb steeply from under 20% at Beginner to over 70% at Advanced.
Trait lift by skill level
Which of these describe you? × How would you describe your skill level with Nix?
Each cell: how over- or under-represented this skill level is among trait-pickers, relative to the overall skill distribution. 1.0× = baseline, >1.0× = over-represented, <1.0× = under-represented. Strips out the population-size imbalance.
Only "rely on tutorials" is over-represented among Beginners (1.30×). Every other trait skews Advanced. Strongest Advanced lifts: "deeper module aspect" (3.34×), "contribute upstream" (3.24×), "architectural choices" (2.79×).
Trait rate by years using Nix
Which of these describe you? × How many years have you been using Nix (at any level of involvement)?
Each cell: of respondents with this much experience, what percent picked this trait. Cells in a column do NOT sum to 100% — traits are multi-choice.
Tutorial reliance starts at 76% in the first year and falls to 26% by 11+ years. Most "advanced" trait rates climb steeply through years 1–6 and largely plateau after. Group sizes shrink rapidly past 5 years (665 → 225 → 98 → 53 → 42), so cells in the rightmost columns are noisier.
Trait lift by years using Nix
Which of these describe you? × How many years have you been using Nix (at any level of involvement)?
Each cell: how over- or under-represented this experience bucket is among trait-pickers, relative to the overall experience distribution. 1.0× = baseline, >1.0× = over-represented, <1.0× = under-represented. Strips out the population-size imbalance.
Tutorial reliance is the only trait over-represented in year-1 respondents (1.20×). All other traits skew toward more-experienced respondents, with strongest lifts at 5+ years for "deeper module aspect" (2.1× → 2.4×) and "contribute upstream" (2.1× → 2.8×). Group sizes shrink rapidly past 5 years, so cells in the rightmost columns are noisier.
Workplace
Workplace adoption
Does your workplace use Nix?
Among respondents whose free-text answers could be normalised, 12.8% reported their workplace uses Nix and 29.2% said it does not. A majority — 57.9% — gave answers that fell into the "Other" category.
Free-text answers normalised to Yes / No / Other.
Workplace decision-maker
Are you the one that made that decision?
Of respondents in a Nix-using workplace, 15.7% report having made the adoption decision themselves; 23.9% did not. The largest share (60.4%) again falls into "Other".
Free-text answers normalised to Yes / No / Other.
Contribution
Contributor experience
If you’ve contributed to Nixpkgs or are interested in contributing, which of the following describe your experience or situation?
Onboarding is the dominant theme. 35.8% plan to contribute in the future, and another 36% want to contribute now but don't know how to get started or how to join existing efforts. Roughly 31% have actually contributed (combining "on my own" and "with community help"). Feedback friction shows up at modest rates: 7.0% report slow feedback, 5.8% got stuck after starting, 2.1% found received feedback unhelpful. Employer policy is rarely a blocker (0.6%). The UpSet chart shows which of these experiences co-occur for the same respondent — each column is one exact combination of selected statements, sized by the bar above. Only the most common combinations are shown; rarer ones (and any with fewer than 5 respondents) are omitted and counted in the chart's caption.
UpSet row key (short code → full statement):
- No plan — I don’t plan to contribute.
- Start? — I would like to contribute, but I don’t know how to get started.
- Join? — I want to contribute to an existing effort, but don’t know how to join in.
- Stuck — I tried to contribute, but got stuck after I started.
- Unclear — I received feedback, but it was unhelpful or unclear.
- Slow — It took too long to receive feedback.
- Blocked — I can’t contribute because my employer doesn’t allow it.
- Covered — Everything I wanted to work on was already being handled by others and I don't feel the need to get involved.
- Solo — I was able to contribute on my own.
- Helped — I was able to contribute with help from the community.
- Future — I plan to contribute in the future.
Foundation
Donation incentives
What would encourage you to donate to the NixOS Foundation?
Transparency is the top motivator: 26.7% would be encouraged by better financial and operational reports — more than any other option. Tax deductibility (16.1%), a maintained LTS release (15.3%), official support or services (12.9%), and donor perks (12.4%) cluster as secondary motivators. Ear-marking donations to specific areas lands last at 5.6%.
Foundation funding priorities
Which areas should the NixOS Foundation prioritise excess funds for?
Each bar shows how respondents distributed their ranks across positions, with the remainder unranked. Documentation is placed first far more often than any other area. Nixpkgs package maintenance and security follow closely. Community health, NixOS releases, and Nixpkgs architecture cluster in the middle. Marketing and community events are the priorities most often ranked low or left unranked.
Information seeking
Objects you interact with
Which objects do you interact with or search for most often across the ecosystem?
Each bar shows how respondents distributed their ranks across grouped positions (1-3, 4-6, 7-10), with the remainder left unranked. Search objects dominate the top band: package search and NixOS option search are placed in the top three far more often than anything else. Configuration examples, manuals, the command-line interface, and development environments form the middle tier. Peripheral objects — job postings, press information, reviews, paid support — are mostly left unranked.
Help resources
When you're looking for help with Nix or NixOS, which resources do you turn to most often? (Rank your top 5. If you usually web search first, pick the resources you click on most.)
Each bar shows the share of respondents who placed a resource at each of their top five positions, with the remainder unranked. The official NixOS Wiki and the reference manuals are placed in the top five most often, followed by GitHub issues and discussions. LLMs already appear in many top-five lists, ahead of nix.dev. Social channels — Reddit, Stack Overflow, Matrix, Discord — trail the official and code-adjacent sources; Mastodon and Twitter/X are rarely chosen for help-seeking.