Everyone who is comparing Anaconda and conda to Astral and uv is missing that the conda ecosystem is language-agnostic while uv is python specific. uv won't help you install gfortran, for example. It is not a replacement, unless you only do python (and use at most common non-python libraries that are available on PyPI).
On the other hand you don't have to use anything associated with Anaconda to use the conda ecosystem. Alternative package managers like mamba and pixi rely on the conda-forge channel instead. Pixi in particular (https://pixi.sh/) is sort-of the uv for the conda ecosystem workflow-wise, and works pretty well if you want that.
I admit I'm not super well versed in what conda's main uses are. Python's whole tooling situation has felt like a nightmare & I've tried to keep far away, but I've had to face it a lot more recently because it's so prevalent for AI. Thankfully uv seems to have done a huge amount of what I need.
Still, if the concern is language-agnostic ways to use tooling, mise (nee rtx) is the 1000 pound gorilla in the room today. Incredibly fast well built Rust based tool that has really massively expanded in scope & offerings, with grace & elegance. I thought it was an asdf replacement, for installing/using toolchains, for .tool-versions files. But it's really grown to be a lot more, capable of letting you isolatedly manage tools it can install from a huge variety of backends (pip, npm, cargo, others). https://mise.jdx.dev/dev-tools/backends/
Hadn't heard about that before, looks interesting. But AFAICS mise focuses on installing tools, not actually all dependencies. It integrates with the languages dependency management tools, which doesn't help you if the language doesn't even have one or the integration wasn't written yet.
I'd rather see more adoption of guix for this purpose. It is a single package manager with a functional approach that allows for introspection of all dependencies (down to the bootstrap toolchains used to build the bootstrap toolchains that build your toolchains, which is something that AFAIK no other package manager except for nix can do), has a fairly large package repository, straightforward locking, actually tested packages, and very easy build recipes (unlike conda-forge...).
If you are in a commercial environment, I can only warn to think that using alternative conda clients will be safe. Condaforge for instance will happily download from the main channel if the recipe requires it. It's pretty hard to make sure this does not happen, best solution is to block access on a network level.
Do you have an example for a package on conda-forge that actually does this? I can only find a vague announcement from 2021 that the "defaults channel is now dropped when building conda-forge packages", as well as statements that the conda-forge repositories are considered incompatible with the defaults channel and having both enabled is an unsupported configuration. Access is blocked on the network level anyway.
That can only happen if you as a user have the 'defaults' channel still configured as available, and conda-forge considers it a user error whenever this happens (the official line is `conda-forge is incompatible with the packages provided in defaults`). Many bug reports are closed simply by telling the user to fix their channel priorities and stop mixing the two
So with a fresh installation of one of those alternative conda clients and no user intervention it won't happen, right? On top of that you can block access on the network level as well.
Sure, when your employees are outside of the corporate network they can still download stuff from the default channels, but in the end it is no different than any other license violation they could do. At least with Anaconda there is a semi-effective fix.
I use Miniforge in a commercial environment and never found a package downloading from the main channel. I'm pretty sure a recipe that does that would be blocked by conda-forge reviewers.
And how many people (count them with fingers) use Conda for anything other than Python? It’s a bloatware. People stopped using Conda because these people kept making the bloat worse.
A single one of the gfortran packages released about 20 hours ago already has 362 downloads, so I think it is safe to say that the non-python users of conda-forge are more than just a handful.
If you don't need it that's fine, no one is stopping you from using PyPI and uv instead. But for some that is not a replacement.
And yes, some things about conda(-forge) could be described as bloated. I particularly dislike the convoluted packaging process.
It is always tradeoffs and deciding based on your own use cases. E.g. if you want to distribute tested packages to users of your software then both conda-forge and PyPI are ill-suited for you. They (and most other package managers) do install-time dependency resolution, so the installation cannot be guaranteed to be tested as working at all. Some package managers do that better, so is conda-forge and PyPI obsolete now?
362 downloads over 20 hours isn't that impressive. Not saying Anaconda isn't great. Don't have much experience with it but I hope the devs get paid for all the value delivered, which I think is largem
Well, it is more than a handful, that was the only point.
You seem to be conflating Anaconda with the conda ecosystem. This package is from conda-forge, which is a community project sponsored by Anaconda but otherwise unrelated to it.
I couldn't care less for Anaconda, but with conda-forge and pixi there is a decent general purpose and language-agnostic package management tool for development purposes in the conda ecosystem.
What does it change? Conda-forge users are users of the conda ecosystem. You suggested there is only a handful of those that use it not for python, I provided a counter point.
Or did I misunderstand you and what you meant with "conda" was either anaconda, or conda-the-software? But then the comment about Python doesn't make much sense.
Heh, looks like they've just pivoted to calling what they do advancing AI and managed to mention AI enough times to get a big new valuation.
Seriously, if you've ever used them before, check out their website now. "Advance AI with Clarity and Confidence", "Simplify, safeguard, and accelerate AI value with open source.", "Millions Rely on Anaconda to Advance Their AI Initiatives"
What does any of that mean? No idea, seems like the actual product is the same conda.
Astral's uv is an impressive, ultra-fast Python package manager that’s rapidly clearing up the pip/virtualenv/poetry mess and setting a new bar for dx. However, Astral is a startup with an unclear path to monetization, whereas Anaconda is a mature company focused on enterprise AI/ML offerings and long-term customer relationships, with conda being just one part of their broader stack. It’s entirely plausible for Anaconda to adopt tools like uv. Comparing Astral and Anaconda directly overlooks their vastly different missions and scales—uv’s technical leap could help unify Python tooling, but Anaconda addresses a different problem.
I'm saying when you compare the resources that have gone in vs. the results that have come out the other side?
Putting these two teams up against one another isn't even fair: its like pushing baby chicks into a pond full of Piranhha fish.
It's not just a package manager: uv and ruff and tye are rapidly becoming an ecosystem. You think they don't have plans for Jupyter?
Google was a tiny company without a monetization strategy. Yahoo being gigantic and "divetsified" just made them a tastier meal for a different breed of competition.
I work with less technical users and the problem with UV is that the installation instructions are slightly more complicated.
For users that just want conda to download python + a bunch of packages and won't ever bother to create environments, anaconda will always be superior.
Now, if UV bundles with a "default python version" with an installer, that may change things.
the other use case is if your env requires compiled binaries, you can't do that with uv. i.e., the Intel MKL package is available on conda but not on pypi. We've also run into this with some cuda related packages.
Where's Anaconda at these days? I've not touched it since starting out with Python and don't think I've touched it since discovering other package managers.
Anyone here using it regularly in 2025, is there anything I'm missing out on?
Everywhere where they may be a shadow data science team. And that's how they get you. One has to be very careful to install the Open Source portions. They make no effort to make the boundary obvious. If one does not, IT will get a call asking for licensing money.
Can confirm, this is exactly what happened. They demand ridiculous back payments unless you buy multiple-year licenses. It would be trivial for them to require an account to use the main channel, but they deliberately make it easy to accidentally use it. If you have to use it, make sure to DNS block anaconda.com (.org seems to be fine), but even better, just avoid them like the plague.
This will also happen if you use condaforge, which can be downloaded freely. Condaforge will also download from the main channel if the recipe requires it.
Setting up a python environment for ML work (pytorch + Nvidia) is simpler with Anaconda, it's a pure dependency nightmare doing it with something else.
Counterpoint: conda is a constant, neverending source of pure dependency nightmares here, that consistently costs us a two-digit percentage of velocity. I'm glad it's working for you, but I really wouldn't generalize. Or recommend it to anyone getting into Python. Use uv, new people, for real.
To be fair, both you and GP are correct. Conda is really really painful to maintain, and it's also much much better for python stuff that relies on native libraries (i.e. C/Fortran/Cuda).
Like, I love uv but GDAL (to use a random recent example) is much easier to install and maintain with conda.
Yeah, I remember finding that the case for a while. I can't remember when things clicked and it was fine without. I moved to arch a while back, perhaps then.
Anaconda sells a managed environment for data science applications, right? Basically the Red Hat business model?
I've used conda for years and haven't set aside the day it'll take to switch to something simple and modern (uv's top of the list, but I'm open to suggestions)
use uv, you dont need to bin your existing installation (it just wont get used anymore) and it will take you 5 minutes to switch. next time you get new hardware you wont bother installing anything else python related.
Today started with coffee and hopes for a new Jon Voight vehicle in theaters next summer. Now it's coffee and disappointment in yet another AI offering no one asked for.
Wait, is this anaconda python, my favorite python installation? Oh no, it's now an AI company? Is my favorite python installation going to get enshittified?
In the meantime, I am volunteering for a non-profit that helps FOSS projects secure sustainable funding, and goodness, that is soooo hard! Enterprises (where money is) are afraid of FOSS, and many prefer to engage commercially with commercial open source companies, backed by VCs
micromamba + conda-forge does the job really well. All open source and community supported, and none of the licensing drama.
In my experience uv (haven't tried astral) doesn't quite fill the same niche, especially if compiled packages from other languages are necessary for your workflow (libboost for example)
Anaconda makes less sense to me, but cursor does have revenue numbers. I haven't seen them so I'm not sure if they look good (we use API keys with cursor so I'm pretty sure they get pure saas margins from us)
I would also venture to guess that cursor is a somewhat nontrivial modification to vscode at this point.
Anaconda != the conda ecosystem != Python
Everyone who is comparing Anaconda and conda to Astral and uv is missing that the conda ecosystem is language-agnostic while uv is python specific. uv won't help you install gfortran, for example. It is not a replacement, unless you only do python (and use at most common non-python libraries that are available on PyPI).
On the other hand you don't have to use anything associated with Anaconda to use the conda ecosystem. Alternative package managers like mamba and pixi rely on the conda-forge channel instead. Pixi in particular (https://pixi.sh/) is sort-of the uv for the conda ecosystem workflow-wise, and works pretty well if you want that.
I admit I'm not super well versed in what conda's main uses are. Python's whole tooling situation has felt like a nightmare & I've tried to keep far away, but I've had to face it a lot more recently because it's so prevalent for AI. Thankfully uv seems to have done a huge amount of what I need.
Still, if the concern is language-agnostic ways to use tooling, mise (nee rtx) is the 1000 pound gorilla in the room today. Incredibly fast well built Rust based tool that has really massively expanded in scope & offerings, with grace & elegance. I thought it was an asdf replacement, for installing/using toolchains, for .tool-versions files. But it's really grown to be a lot more, capable of letting you isolatedly manage tools it can install from a huge variety of backends (pip, npm, cargo, others). https://mise.jdx.dev/dev-tools/backends/
Hadn't heard about that before, looks interesting. But AFAICS mise focuses on installing tools, not actually all dependencies. It integrates with the languages dependency management tools, which doesn't help you if the language doesn't even have one or the integration wasn't written yet.
I'd rather see more adoption of guix for this purpose. It is a single package manager with a functional approach that allows for introspection of all dependencies (down to the bootstrap toolchains used to build the bootstrap toolchains that build your toolchains, which is something that AFAIK no other package manager except for nix can do), has a fairly large package repository, straightforward locking, actually tested packages, and very easy build recipes (unlike conda-forge...).
If you are in a commercial environment, I can only warn to think that using alternative conda clients will be safe. Condaforge for instance will happily download from the main channel if the recipe requires it. It's pretty hard to make sure this does not happen, best solution is to block access on a network level.
Do you have an example for a package on conda-forge that actually does this? I can only find a vague announcement from 2021 that the "defaults channel is now dropped when building conda-forge packages", as well as statements that the conda-forge repositories are considered incompatible with the defaults channel and having both enabled is an unsupported configuration. Access is blocked on the network level anyway.
That can only happen if you as a user have the 'defaults' channel still configured as available, and conda-forge considers it a user error whenever this happens (the official line is `conda-forge is incompatible with the packages provided in defaults`). Many bug reports are closed simply by telling the user to fix their channel priorities and stop mixing the two
Correct, it's a user error, but in a corporate environment, this happens. Many scientists have their own recipes and you can't catch them all.
So with a fresh installation of one of those alternative conda clients and no user intervention it won't happen, right? On top of that you can block access on the network level as well.
Sure, when your employees are outside of the corporate network they can still download stuff from the default channels, but in the end it is no different than any other license violation they could do. At least with Anaconda there is a semi-effective fix.
I use Miniforge in a commercial environment and never found a package downloading from the main channel. I'm pretty sure a recipe that does that would be blocked by conda-forge reviewers.
And how many people (count them with fingers) use Conda for anything other than Python? It’s a bloatware. People stopped using Conda because these people kept making the bloat worse.
A single one of the gfortran packages released about 20 hours ago already has 362 downloads, so I think it is safe to say that the non-python users of conda-forge are more than just a handful.
If you don't need it that's fine, no one is stopping you from using PyPI and uv instead. But for some that is not a replacement.
And yes, some things about conda(-forge) could be described as bloated. I particularly dislike the convoluted packaging process.
It is always tradeoffs and deciding based on your own use cases. E.g. if you want to distribute tested packages to users of your software then both conda-forge and PyPI are ill-suited for you. They (and most other package managers) do install-time dependency resolution, so the installation cannot be guaranteed to be tested as working at all. Some package managers do that better, so is conda-forge and PyPI obsolete now?
362 downloads over 20 hours isn't that impressive. Not saying Anaconda isn't great. Don't have much experience with it but I hope the devs get paid for all the value delivered, which I think is largem
Well, it is more than a handful, that was the only point.
You seem to be conflating Anaconda with the conda ecosystem. This package is from conda-forge, which is a community project sponsored by Anaconda but otherwise unrelated to it.
I couldn't care less for Anaconda, but with conda-forge and pixi there is a decent general purpose and language-agnostic package management tool for development purposes in the conda ecosystem.
See, why are you adding forge in there? Why do you think forge exists?
What does it change? Conda-forge users are users of the conda ecosystem. You suggested there is only a handful of those that use it not for python, I provided a counter point.
Or did I misunderstand you and what you meant with "conda" was either anaconda, or conda-the-software? But then the comment about Python doesn't make much sense.
Heh, looks like they've just pivoted to calling what they do advancing AI and managed to mention AI enough times to get a big new valuation.
Seriously, if you've ever used them before, check out their website now. "Advance AI with Clarity and Confidence", "Simplify, safeguard, and accelerate AI value with open source.", "Millions Rely on Anaconda to Advance Their AI Initiatives"
What does any of that mean? No idea, seems like the actual product is the same conda.
Going up against Astral in 2025 with conda's stack is feeding 150MM directly into a furnace.
Astral's uv is an impressive, ultra-fast Python package manager that’s rapidly clearing up the pip/virtualenv/poetry mess and setting a new bar for dx. However, Astral is a startup with an unclear path to monetization, whereas Anaconda is a mature company focused on enterprise AI/ML offerings and long-term customer relationships, with conda being just one part of their broader stack. It’s entirely plausible for Anaconda to adopt tools like uv. Comparing Astral and Anaconda directly overlooks their vastly different missions and scales—uv’s technical leap could help unify Python tooling, but Anaconda addresses a different problem.
I'm saying when you compare the resources that have gone in vs. the results that have come out the other side?
Putting these two teams up against one another isn't even fair: its like pushing baby chicks into a pond full of Piranhha fish.
It's not just a package manager: uv and ruff and tye are rapidly becoming an ecosystem. You think they don't have plans for Jupyter?
Google was a tiny company without a monetization strategy. Yahoo being gigantic and "divetsified" just made them a tastier meal for a different breed of competition.
$50m more than Oxide Computer, a company actually building something, just raised. For what exactly? A free python distro, doing some vague AI pivot?
For people asking who uses Anaconda nowadays.
I work with less technical users and the problem with UV is that the installation instructions are slightly more complicated.
For users that just want conda to download python + a bunch of packages and won't ever bother to create environments, anaconda will always be superior.
Now, if UV bundles with a "default python version" with an installer, that may change things.
the other use case is if your env requires compiled binaries, you can't do that with uv. i.e., the Intel MKL package is available on conda but not on pypi. We've also run into this with some cuda related packages.
Where's Anaconda at these days? I've not touched it since starting out with Python and don't think I've touched it since discovering other package managers.
Anyone here using it regularly in 2025, is there anything I'm missing out on?
Everywhere where they may be a shadow data science team. And that's how they get you. One has to be very careful to install the Open Source portions. They make no effort to make the boundary obvious. If one does not, IT will get a call asking for licensing money.
Can confirm, this is exactly what happened. They demand ridiculous back payments unless you buy multiple-year licenses. It would be trivial for them to require an account to use the main channel, but they deliberately make it easy to accidentally use it. If you have to use it, make sure to DNS block anaconda.com (.org seems to be fine), but even better, just avoid them like the plague.
How are they going to find out though?
They have a form where they ask for an email before giving the download link. My guess is that + a bit of telemetry.
This will also happen if you use condaforge, which can be downloaded freely. Condaforge will also download from the main channel if the recipe requires it.
You can bypass that
By checking IP, I'm guessing
I work from home and even if I didn't, how are they going to link an IP to a company
Setting up a python environment for ML work (pytorch + Nvidia) is simpler with Anaconda, it's a pure dependency nightmare doing it with something else.
Counterpoint: conda is a constant, neverending source of pure dependency nightmares here, that consistently costs us a two-digit percentage of velocity. I'm glad it's working for you, but I really wouldn't generalize. Or recommend it to anyone getting into Python. Use uv, new people, for real.
To be fair, both you and GP are correct. Conda is really really painful to maintain, and it's also much much better for python stuff that relies on native libraries (i.e. C/Fortran/Cuda).
Like, I love uv but GDAL (to use a random recent example) is much easier to install and maintain with conda.
Yeah, I remember finding that the case for a while. I can't remember when things clicked and it was fine without. I moved to arch a while back, perhaps then.
Docker
Anaconda sells a managed environment for data science applications, right? Basically the Red Hat business model?
I've used conda for years and haven't set aside the day it'll take to switch to something simple and modern (uv's top of the list, but I'm open to suggestions)
use uv, you dont need to bin your existing installation (it just wont get used anymore) and it will take you 5 minutes to switch. next time you get new hardware you wont bother installing anything else python related.
Heh, like I'm still holding on to pip. uv looks fab, really need to give it a go.
Just add 'uv' in front of your pip commands, that's how it's called anyway
Anaconda has 150mm in annual recurring revenue? That's excellent news. It'll be interesting to see how this investment helps them grow.
> This news comes on the heels of Anaconda’s newly launched AI Platform
Ahh makes sense now.
The efficient hand of capitalism at work once again allocating capital to the most effective endeavours.
> the most effective endeavours
Slapping Ai somewhere, anywhere on every page of the pitch deck is the most effective endeavour of the 2020's
Today started with coffee and hopes for a new Jon Voight vehicle in theaters next summer. Now it's coffee and disappointment in yet another AI offering no one asked for.
150M from people who never used Anaconda...
I cannot imagine using Anaconda with how many issues I had. Virtual Environments have been superior.
Wait, is this anaconda python, my favorite python installation? Oh no, it's now an AI company? Is my favorite python installation going to get enshittified?
That happened last year, when they changed their license terms and started shaking people down for money. I wouldn't touch it.
https://www.theregister.com/2024/08/08/anaconda_puts_the_squ...
https://licenseware.io/retrospective-on-anacondas-2024-licen...
https://www.cdotrends.com/story/4173/anaconda-threatens-lega...
https://www.reuters.com/legal/litigation/intel-sued-copyrigh...
The same story again and again...
In the meantime, I am volunteering for a non-profit that helps FOSS projects secure sustainable funding, and goodness, that is soooo hard! Enterprises (where money is) are afraid of FOSS, and many prefer to engage commercially with commercial open source companies, backed by VCs
Last year is just when they started being honest about it. Before then they left their terms intentionally ambiguous, then shook people down anyway.
Good eye. I would not have noticed if the terms changed,even with all of those articles. Is there a good replacement?
micromamba + conda-forge does the job really well. All open source and community supported, and none of the licensing drama.
In my experience uv (haven't tried astral) doesn't quite fill the same niche, especially if compiled packages from other languages are necessary for your workflow (libboost for example)
uv.
Just replace it with Miniforge: https://conda-forge.org/download/
Or mamba/micromamba as well
Congratulations to Astral (makers of uv).
At these valuations you are now all billionaires. You have earned the commas and car doors that open like \__/.
Unfortunately they've made the mistake of calling their Python package manager a Python package manager, rather than the secret to advancing AI.
What is their revenue?
They're a pure pre revenue play.
Do you think Anaconda founders will ever become billionaires? If not, why would you expect it for Astral founders?
Anaconda just raised funds at a valuation of 1.5B so they are well on their way.
other than having "AI" in the name, I didn't see anything about its "AI Platform" that's actually AI in a meaningful way
To be fair, the easiest way to maintain dependencies for computing on GPUs is defintely conda. Anything else requires a bunch more manual toil.
Is it just me or is trivial software getting overfunded? How does an AI skin on vscode or conda have a big enough moat for these crazy rounds?
Anaconda makes less sense to me, but cursor does have revenue numbers. I haven't seen them so I'm not sure if they look good (we use API keys with cursor so I'm pretty sure they get pure saas margins from us)
I would also venture to guess that cursor is a somewhat nontrivial modification to vscode at this point.
[dead]