At first I thought people here were being pretty unsympathetic to an early version of a beneficial program. I could see a company setting a 6-month timeline initially, so they can reevaluate the program and choose how to evolve their support for open source. I expected to see something along the lines of, "at the end of the 6 months we'll evaluate whether to continue your free plan."
But no, they're quite explicit about this being nothing more than a way to try to get paid subscriptions from open source maintainers:
> Your complimentary subscription will expire at the end of the Benefit Period. After expiration, any existing subscription will continue unless you cancel. You may independently choose to purchase a paid Claude subscription at the then-current price through Anthropic’s standard signup process.
So anyone who participates in this will need to remember to opt out six months from now, or suddenly find themselves with invoices at the max 20x level.
That's pretty ugly.
Edit: I believe I misread the terms. As mwigdahl points out below: "If you have an existing subscription, it pauses while the free period is active. After that free period, your existing subscription resumes. As I read it, there is no "auto-subscribe" after the free period ends -- you just revert back to whatever you had before (or nothing, if you weren't a subscriber before)."
This does not appear to be true if you read the earlier "Activation" section. If you have an existing subscription, it pauses while the free period is active. After that free period, your existing subscription resumes. As I read it, there is no "auto-subscribe" after the free period ends -- you just revert back to whatever you had before (or nothing, if you weren't a subscriber before).
Even if they did let the free users continue using, and then preesnted them with invoices, those would mean nothing without a registered, up-to-date payment method on file.
Tons of SaaS companies offer open source projects free periods or a limited hobby plan for free. Claude is offering a professional plan 20x'd for a free period. I don't see anything wrong with that. This is a far more resource expensive service to offer for free than 99% of SaaS companies.
This does not strike me as an anti-pattern or ugly. Indefinite free period would be unreasonable, and automatically kicking a user off would also probably be bad. A $200 bill shock is not great but it's also at a size that won't cause enormous distress while simultaneously being noticeable enough that you won't pay more than a month over. (As an open-source maintainer already on a Max plan, I still wince every month.) Income-constrained users should not adopt it or should set a reminder well beforehand.
Your suggestion of "we'll evaluate" individually would be a very costly undertaking for Anthropic. Not reasonable. If your suggestion was for Anthropic to evaluate at the end of the 6 months whether to continue the free plan generally, I don't see anything that prevents them from doing so.
I think Anthropic should probably give some notice in the CLI or Claude.ai in the final month of the offer. Not doing that would be a bit ugly.
> automatically kicking a user off would also probably be bad.
No. "Sorry, subscription has expired, please re-up your account" is an extremely reasonable UX.
The whole "free period but we'll auto bill you after" is a shitty dark pattern that mostly exists to extract value from life admin errors. The people who got enough value to justify the cost would've paid anyway.
A $200 bill from some cloud entity that doesn't have my credit card info would cause nothing but enormous laughter.
What is ugly here is the combination of the free trial (not ugly in an of itself), and they way they are trying to recruit qualified users for it from open source.
To be honest, it's quite likely that someone who applies is already paying $20/month and would save them for 6 months, so the extra shock is only $60. And it's quite easy to set up a calendar event to remember to unsubscribe.
I have had subscriptions renewed unwillingly and it was always clear to me that, as much as I disliked this practice, the expense was always my fault.
Ugly may be a strong word, but upon reading the title, the first thought that came to me was that they'd done some self-examination and decided to finally do the ethical thing about all the open source training data without which their proprietary product would plain and simply not exist.
In comparison, a program that grants time-limited credits to a few high-visibility projects reads like a self-serving marketing move no matter how you slice it.
What baffles to me is the people who think that "gifts" should never be criticized.
I mean, suppose Adobe decides to gift "$1200" value in Adobe products/subscriptions to all subscribers of the gimp-users mailing list. Can I criticize that?
My calendar is littered with the occasional "Cancel Wired subscription", "Cancel Amazon Unlimited", "Cancel Fitbit premium". This is a standard promotional offer, and it's trivial to not get bitten by it. We have the technology to set reminders for future dates.
It's not trivial for me. All my life I've struggled to attend to scheduled events that are not regularly recurring. I've missed midterm exams in college. I've missed band gigs I was scheduled to play in. I've accidentally stood people up in social outings. I've missed credit card payments. (solved that one with auto-pay) I have calendars and email accounts, and they usually work, but sometimes I miss the notification or forget to check the calendar.
For me, if I was going to plan to cancel something in the future, then instead of scheduling it, I'd just do it now before the thought goes out of my head.
It is disgusting. I just use "fake" credit cards from online services to end-around this. Obnoxious for sure, but it saves me the headache of tracking this kind of shit.
That never works for me. I try to only sign up for things that I can cancel immediately and continue to use for the rest of whatever time period I signed up for.
Instead of potentially getting billed for some trial I forgot about, I would rather pay for a month, immediately cancel, and then repeat every month when I realize it's not working.
Besides helping me keep my expenses under control, it doubles as an evaluation of the company. If they make it difficult to cancel, or do not let me use the rest of my paid time, I know they are not a company I want to do business with.
You're absolutely right that some individuals will be able to sign up for this program, and remember to cancel at the end of the six months. However, when companies choose to implement a policy like this they're acting on well-established statistics. They know that a meaningful percentage of people will forget to cancel, and the company will end up with increased revenue. There might be a bit of good will here, but in the end a program like this with these clearly-spelled-out terms is not much more than marketing.
This feels especially ugly to me because maintainers of large open source projects will feel pressure to keep using tools that let them work in an AI-assisted world. This really feels like it will make life harder for open source maintainers in the end, rather than easier. That's the opposite of what a meaningful open source campaign should look like.
At the very least, it puts maintainers right back in the position of having to beg giant companies for handouts.
It seems like the average payoff is not so relevant if you have good reason to believe you can do better than average. Also, I'm not so sure Anthropic would profit from this particular offer in the average case.
I recently downgraded from Opus to Sonnet because it's 40% cheaper and it needs a bit more guidance but seems doable. There will likely be better deals.
OSS maintainer: I'd like to cancel my subscription!
Claude: Thank you for prolonging your subscription for another year. I'll take the required steps.
OSS maintainer: No, I said CANCEL!
Claude: You are absolutely right! Thank you for your two year subscription.
It should be a reason to criticize them, though. They're tricking people in order to make more money. They know it, you know it, we all know it. They could easily not do this, or if they want to make the argument that it's helpful not to have your subscription suddenly lapse at the end of the period, they could make it an option to have your subscription auto-renew as paid.
> You’re a primary maintainer or core team member of a public repo with 5,000+ GitHub stars or 1M+ monthly NPM downloads.
I've been an open source maintainer of one of the biggest open source projects in the world[1], and it wouldn't fill any of these requirements. Anybody else hates it that now "open source" is conflated with Github (a private company, itself not open source) popularity?
Maybe worth asking for anyway? They might just be setting metrics based on the most popular ways of measuring but if they care about the spirit of the offer it would make sense for them to be flexible with the letter of the requirements.
Considering they trained their model on open-source software, the least they could do is give it to open-source maintainers for free with no time limit. I’m sure they can come up with other ways to prevent abuse. This 6-months-free move just adds insult to injury, like it’s just a move to extract more from those who involuntarily contributed to the training already. And that’s coming from me, a Claude Code fan.
The double standards are so obnoxious. Corporations bent over backwards to lobby intellectual property into law, then they invent AI and suddenly everything turns into fair use.
> Considering they trained their model on open-source software, the least they could do is give it to open-source maintainers for free with no time limit.
Why? The resulting code generated by Claude is unfit for training, so any work product produced after the start of the subsidized program should be ignored.
Therefore it makes sense to charge them for the service after 6 months, no? Heh.
What do you mean it's unfit for training? It's a form of reinforcement learning; the end result has been selected based on whether it actually solved the need.
You need to be careful of the amount of reinforcement learning vs continued pretraining you do, but they already do plenty of other forms of reinforcement learning, I'm sure they have it dialed in.
Anthropic’s models have almost certainly gorged on an enormous amount of OSS, and if they think they can settle that debt with only six months of perks for the maintainers who’ve kept that ecosystem alive, it comes across as pretty arrogant.
It's amazing how quickly Anthropic is turning into the "bad" guys.
First we couldn't use our Claude subscription with anything but Claude code, then the limits seemed to change every week without any communication, then they banned a bunch of people (including some prominent names). Then they complain about the Chinese distilling using their API (which I'm partly sympathetic to but let's not pretend that Antrophic invented their training data from scratch).
Then there's this half-baked offer. I mean sure, it looks nice on paper but given how incredibly valuable opensource has been for them and given their budget it does seem a bit tight.
Uncharitably, I think this is a strategy to gorge further especially if they select for higher quality open source. They are embracing the best to train off iteration patterns of the best, and have a semi self correcting slop mechanism.
Charitably this will be great for open source software so... so long as they never moat up and lockdown.
I like what GitHub and Jetbrains are doing, where you get Copilot and PyCharm for free as long as you're a maintainer. They keep renewing my license.
A 6-month trial isn't showing appreciation for OSS any more than "first crack hit's free" is showing appreciation for what a good person you are. It's just "you look like a promising customer".
It would be showing greater higher quality appreciation to offer an ongoing benefit.
But there is some benefit to giving maintainers a generous trial length with your offering. 6 months is certainly long enough to see how well it does or does not incorporate into your project.
It just so happens we almost all universally love the offering.
> But there is some benefit to giving maintainers a generous trial length with your offering. 6 months is certainly long enough to see how well it does or does not incorporate into your project.
This would be fine in the context of a general sales pitch/marketing deal.
But OSS development and maintenance is special here. It has a budget of $0. As a sales strategy, Anthropic would be better off trying to sell luxury gold plated bindles to hobos.
And there's another question: How exactly does Anthropic see the future of OSS, with this pitch? What are they thinking? Is this the new norm for OSS a $200/month entry fee?
Because adding such a cost to OSS would not only go against everything OSS stands for, and would push the vast majority of maintainers into quitting their projects.
(Now, Anthropic can't mandate maintainers use Claude, though a much-discussed side effect of tools like Claude has been the increased burden on OSS maintainers. And while Anthropic does not raise suggestion that they deal with this by employing AI tools, bystanders most certainly have.)
Github gives Copilot Pro to open source maintainers but they don't really tell you what the requirements are.
I have it and I just get a notification every month that it's been renewed and I never even applied for it.
I assume it's a combination of activity on github and popularity of your repositories.
I mean, there's also the whole GitHub free tier. It used to only be for public repos, so mostly OSS plus "shared source", but now they allow it for private as well. But it still costs them money to host your code and provide CI minutes.
It's weird to make it 6 months only because it sends a message of, "Thank you for dedicating 5-10+ years building up a very popular open source project. In return we believe this is worth exactly $1,200 (6 x $200) in credits". Especially since they are scraping all of our work and profiting from it directly without acknowledgement or compensation -- past, present and future indefinitely.
I don't get these negative comments for them giving free credits. Either it's "Not many people fit these criteria" or "A ploy for more customers". It can't be both, and I believe it's neither. It's a nice gesture, in line with Github Copilot and JetBrains. Disclosure: I have free Copilot and just applied for 20x Pro.
Made a mistake reading this thread on Safari where I don't have the usual suspects blocked. Some guy read that this converts to paid and then a bunch of people just kept repeating it. A real lesson in how many people are simply repeating things without knowing anything.
> Maintainers: You’re a primary maintainer or core team member of a public repo with 5,000+ GitHub stars or 1M+ monthly NPM downloads. You've made commits, releases, or PR reviews within the last 3 months.
How many total developers does that cover? 100? How many of them aren't already corporate employees?
And also
> 6 months of free Claude Max 20x
So basically a free trial.
When Github Copilot first launched they gave Pro subscriptions to everyone that regularly committed to a public repo, regardless of the number of stars or downloads, and kept renewing it indefinitely. I don't know if that program is still around but it was amazing to get to try out some early LLM coding tools for open source development.
Github search gives me 11 300 results for 5000+ stars[0]. Dunno if they all qualify as open source, but that's also repos, not contributors. Presumably there's an average of > 1 per repo.
NPM probably adds a lot. I can't find any recent sources, but NPM packages get downloaded a lot (e.g., every Github Action run.) And to get such a download, an NPM package just has to be somewhere in the dependency tree, which are famously enormous. (Though many might not be updated in the past 3 months, though.)
GitHub is cagey about the criteria, but yes this is ongoing. It doesn't appear to be tied to active contributions though. I'm a maintainer on paper of a moderately large open source project that I haven't been involved with in years, and they still renew my free copilot monthly.
I think there's plenty of them. I know at least 3 guys eligible for such requirements (but this guys aren't some public persons giving tech-talks and so on, just some niche libs for others to use). If Claude would ask for 100k stars repos, then yeah. I guess there would be even less than 100
Yeah, their thing is more making products worse over time and wasting billions. You will see this in action shortly with XBox. I think they will do both this time.
This is going to get abused so fast, it will make your head spin.
EDIT: I just look up the highest-ranking "buy GitHub stars" page (which I will obviously not link here), and it looks like you would have to pay a little over $1000 to get the required amount of stars. So I suppose it might not get abused as easily as I thought.
On the other hand, someone with the gumption and elbow grease to abuse this process themselves could still easily do so, I'd wager.
All that being said, I still think that GitHub stars are effectively worthless, and attempting to assign value to them like this is, at best, a fool's errand.
I can imagine this will invoke Goodhart's law, increasing the amount of people shilling their AI-generated shovelware onto a Web already greatly suffering from the consequences of the plummeting cost of intelligent-sounding text generation.
They do require that you allow them to use your name publicly.
They are silent on whether you can prohibit them from training on your input, so I assume you can.
My guess is, if even 10% of maintainers forget to disable training, then Anthropic will have a most excellent source of how really good developers approach problems that can be fed back into the model. That could improve things for everyone.
Of course, the whole purpose of a trial is to induce dependence on the service. Let’s hope that doesn’t reduce the skill of those maintainers. If open source code gets better as a result, that’s good for all.
AI is somewhat helpful but I'm not interested in a company finding a way for me to pay to do my volunteer OSS work. GitHub Copilot offers a permanent free subscription for OSS maintainers.
I previously ignored a free offer when Claude reached out to me as an open source maintainer as it was a glorified free trial. I hope this one continues beyond the listed 6 months, I am not interested in a glorified free trial and if it requires entering credit card details I won't be signing up.
Open source developers should be paid for their efforts, and for their contributions to LLM models, past, present, and future, rather than be enticed into paying to participate six months down the road.
OSS developers driven by something else than just money I believe. They are proud of their work of giving something to the community with their name on it. So such respect as giving free subscription to them I think matters, as they were mentioned and respected.
Sorry we stole all your src code that you labored over for hours and hours of your life. Here’s a few bucks for 6 months to help train our model even more.
>Maintainers: You’re a primary maintainer or core team member of a public repo with 5,000+ GitHub stars or 1M+ monthly NPM downloads. You've made commits, releases, or PR reviews within the last 3 months.
It converts back to paid automatically if you had an existing paid subscription before. No other cases. In any case, this is still a valuable service they are providing for 6mo for free, which many will appreciate even if the goal is to recruit more users.
I get Copilot for free as an open source maintainer and it's nice. But right now I am also paying for two Claude Max ($200/mo) for my own projects. Would be nice to have one of them covered for at least 6 months! Hope Anthropic accepts my application because I do not track downloads at all.
I really appreciate the gesture, but this kind of feels like it’s an attempt to claw (lol) some good will back from devs. The barrier is way too high, imo. And the 6 month cap does make sense given the cost of LLMs but it’s a bad feeling. We like you, but for 6 months.
As a tinnnyy plug, I’ve ran OSS sponsorship programs before for companies. One thing that I always hated was the sales contact process to get it. So, for Vizzly I made it 100% automated. Sign up, connect an OSS public repo, get a free plan. https://vizzly.dev/open-source/ I don’t wanna talk to you and you don’t wanna talk to me (for this :p)
Gotta boost up those user numbers before dumping the stock in an IPO, I'd respect it more if it wasn't what every other tech IPO has done over the previous decade.
5000 stars. That's an interesting threshold.
I've checked and astropy -- the main python module used by pretty much every python user in astrophysics has 5100 stars. I would guess almost no open source code in science would pass the threshold.
EDIT: Just another test, one of the most used codes in astro -- an ensemble Markov-Chain Monte-Carlo sampler https://github.com/dfm/emcee has 1600 stars.
It just shows the 5000 stars is a bit PR, rather than a serious attempt to help open source.
Yeah, I was going to come here to say this. Apart from a) stars are a dubious metric b) 5000 stars is an insanely high bar, there is the issue that there are definitely lots of projects that choose not to partake in GitHub at this point.
That said, they do have a "contact us" line in there which implies some flexibility.
You can easily buy stars in bulk, like you can buy social media “likes” so they are kind of measuring the wrong thing and incentivizing the wrong behaviors.
I don't even star the vast majority of packages I use... I usually only star repos I don't use but find interesting and may want to refer back to in the future.
And completely excludes projects not hosted on Microsoft's GitHub or NPM (Though they do say you can contact them if you don't meet their insane criteria).
I may sound unthankful here, but it just very strongly smells of Antropic amping up their PR campaigning lately, even the headline on the post reads offputting.
Plus, while 6 months is better than 1 month, why isn't it a recurring deal (or token-limited), which renews after check-ins (like educational discounts do). This sounds like an Apple TV+ offer you get for every Apple product you buy. A hook, more than a treat.
In this case, I guess it's just a slimy approach to building a self-selected lead list of people you can hard-hit with upsells after the 6 months.
I'm guessing that this is an initial trial and they're intending to extend it further; 6 months is a reasonable trial period given the very rough metric for deciding who qualifies.
> Your complimentary subscription will expire at the end of the Benefit Period. After expiration, any existing subscription will continue unless you cancel. You may independently choose to purchase a paid Claude subscription at the then-current price through Anthropic’s standard signup process.
Sure. They're still figuring out exactly how to decide who qualifies for this -- seriously, 5000 github stars or 1M monthly NPM downloads? -- and they don't want to make long-term promises to people who might not qualify under future better-thought-out rules.
That doesn't mean they're not going to continue this, it just means they're being careful not to make promises which they'll want to roll back later.
No, thanks. I decided I don't want to play those games. I get MiniMax unlimited for 10$ per month, and free GitHub Copilot as an open source maintainer and contributor.
I don't need to beg to get some free stuff, only to later realize the only way to use it is through the shitty Claude Code.
I'm kind of sceptical about the altruistic motives here. Giving this to open source maintainers also solves the problem of identifying high quality feedback/rewards for their rlvr models. With everybody using Claude code it might be difficult for them to find a robust way to tell apart good reward signal from mediocre or below average feedback.
I'll take it! I've been using Opus 4.6 with GitHub Tasks sparingly but any sort of continued usage is very expensive. This would be handy, like 10x my efforts.
Now suddenly everyone's gonna become a 'maintainer'.
People are gonna abuse it and just use it for everything else BUT proper(not fake and AI GENERATED) open source projects.
Sad day. I hope so they are gonna change the TOS and punish anyone with a 1 million $ fine if someone lies.
That's the only way: criminal charges for students using AI(when forbidden such as academia) and people who plan to abuse it (stealing tokens against TOS).
it's impossible to compete with cheaters and with cheaters who stole money
If you appreciate open source maintainers, detect when users are opening pull requests without human review and stop them. Feel free to keep burning their tokens, just stop making pull requests.
Yeah, I think a lot of open-source maintainers would rather have some kind of an anti-slop filter than a six-month trial. All of my GitHub projects are tiny so I haven't had to encounter it, but I've heard that some projects are absolutely swamped in crap.
In the past week (besides the constant slop), there are models which have misattributed the copyright of new files to me, and stripped my copyright from existing files. It's sapping up time, energy and motivation.
Hey that seems pretty cool! No doubt it's gonna be a way to either collect more info of successful devs or maybe just upsell stuff after those 6 months are over, but it's something!
I went for their 100 USD paid tier and it's honestly been immensely useful (Claude Code with the desktop UI with multiple parallel tasks), I've done more and with better quality in the past few weeks than others do in a month - maybe I just got lucky with the domain but it really is a force multiplier and I'm working on like 4 projects in parallel at work and am crushing it, being overworked aside.
Finally I also have enough capacity for various side projects and utility tools/scripts, or at least I will until I burn out, but that's not really the fault of the tool, rather the amount of work.
Being able to throw the latest Opus model at every problem is also really, really nice. Way better than any of the slop before.
It also makes sense to give tools for open source developers. Sometimes we need to test compatibility (does my repo play nice with that harness/ide/etc?). This in turn makes that repo be more solid for the paid tool, which is a potential way of attracting users for both. It has been done by others (like JetBrains IDEs).
The cynicism here is crazy. You can get a lot done in 6 months and prices will probably have dropped by then due to competition. There's no lock-in keeping you from switching coding agents if you're not stupid about it.
There's nothing wrong with taking advantage of limited offers.
But the application form isn't asking for credit card info. (Does anyone know whether they ask for that later in the approval process?)
In any case, the fine print says that participants have to purchase after the expiration of the free period in order to to continue. Nothing is mentioned about having to give payment info upfront, such that the account automatically transitions to payment.
Participants who are already paying customers will have their payments suspended for that period, so I think for them it will automatically lapse back to paid, at least if their payment method is up-to-date.
> Maintainers: You’re a primary maintainer or core team member of a public repo with 5,000+ GitHub stars or 1M+ monthly NPM downloads. You've made commits, releases, or PR reviews within the last 3 months.
Laughable.
This is a tiny, if even unimportant, fraction of the FOSS community that runs the modern tech stack.
5000 stars required? And six months only? What a misleading multilevel clickait scam. But I knew that everything about Anthropic is a scam, from the excessive token usage to the model quality reduction to the various user-hostile actions.
No thanks, projects are too important for slop. And why would I want to be tracked so you can see my thought process, stupid questions etc.? Will you sell that information later?
Your CEO has bragged multiple times how your tool will make me unemployed. Why would I participate in that?
You stole my code without attribution. Why should I use the services of a copyright infringer?
Now suddenly everyone's gonna become a 'maintainer'.
People are gonna abuse it and just use it for everything else BUT proper(not fake and AI GENERATED) open source projects.
Sad day. I hope so they are gonna change the TOS and punish anyone with a 1 million $ fine if someone lies.
That's the only way: criminal charges for students using AI(when forbidden such as academia) and people who plan to abuse it (stealing tokens against TOS).
it's impossible to compete with cheaters and with cheaters who stole moneyl
But no, they're quite explicit about this being nothing more than a way to try to get paid subscriptions from open source maintainers:
> Your complimentary subscription will expire at the end of the Benefit Period. After expiration, any existing subscription will continue unless you cancel. You may independently choose to purchase a paid Claude subscription at the then-current price through Anthropic’s standard signup process.
So anyone who participates in this will need to remember to opt out six months from now, or suddenly find themselves with invoices at the max 20x level.
That's pretty ugly.
Edit: I believe I misread the terms. As mwigdahl points out below: "If you have an existing subscription, it pauses while the free period is active. After that free period, your existing subscription resumes. As I read it, there is no "auto-subscribe" after the free period ends -- you just revert back to whatever you had before (or nothing, if you weren't a subscriber before)."
https://www.anthropic.com/claude-for-oss-terms
If I'm reading it wrong, let me know.
I mean, pay this invoice ... or else what?
Or else they send it to collections.
Your suggestion of "we'll evaluate" individually would be a very costly undertaking for Anthropic. Not reasonable. If your suggestion was for Anthropic to evaluate at the end of the 6 months whether to continue the free plan generally, I don't see anything that prevents them from doing so.
I think Anthropic should probably give some notice in the CLI or Claude.ai in the final month of the offer. Not doing that would be a bit ugly.
Would it? The only way to access Claude is via a CLI or a GUI.
> $ claude --resume
> No subscription active (expired on 6/1/2026). Reactivate at claude.ai/settings.
No. "Sorry, subscription has expired, please re-up your account" is an extremely reasonable UX.
The whole "free period but we'll auto bill you after" is a shitty dark pattern that mostly exists to extract value from life admin errors. The people who got enough value to justify the cost would've paid anyway.
What is ugly here is the combination of the free trial (not ugly in an of itself), and they way they are trying to recruit qualified users for it from open source.
I have had subscriptions renewed unwillingly and it was always clear to me that, as much as I disliked this practice, the expense was always my fault.
I think it’s reasonable to grant humans agency. If they don’t want it they don’t have to take it. It’s pretty obviously a huge net positive.
In comparison, a program that grants time-limited credits to a few high-visibility projects reads like a self-serving marketing move no matter how you slice it.
I mean, suppose Adobe decides to gift "$1200" value in Adobe products/subscriptions to all subscribers of the gimp-users mailing list. Can I criticize that?
For me, if I was going to plan to cancel something in the future, then instead of scheduling it, I'd just do it now before the thought goes out of my head.
Instead of potentially getting billed for some trial I forgot about, I would rather pay for a month, immediately cancel, and then repeat every month when I realize it's not working.
Besides helping me keep my expenses under control, it doubles as an evaluation of the company. If they make it difficult to cancel, or do not let me use the rest of my paid time, I know they are not a company I want to do business with.
This feels especially ugly to me because maintainers of large open source projects will feel pressure to keep using tools that let them work in an AI-assisted world. This really feels like it will make life harder for open source maintainers in the end, rather than easier. That's the opposite of what a meaningful open source campaign should look like.
At the very least, it puts maintainers right back in the position of having to beg giant companies for handouts.
I recently downgraded from Opus to Sonnet because it's 40% cheaper and it needs a bit more guidance but seems doable. There will likely be better deals.
It may or may not be worth playing their game depending on whether you use the product or not, but there are opportunities for people who do play.
I've been an open source maintainer of one of the biggest open source projects in the world[1], and it wouldn't fill any of these requirements. Anybody else hates it that now "open source" is conflated with Github (a private company, itself not open source) popularity?
[1]: https://www.openstack.org/
> Don't quite fit the criteria If you maintain something the ecosystem quietly depends on, apply anyway and tell us about it.
Why? The resulting code generated by Claude is unfit for training, so any work product produced after the start of the subsidized program should be ignored.
Therefore it makes sense to charge them for the service after 6 months, no? Heh.
You need to be careful of the amount of reinforcement learning vs continued pretraining you do, but they already do plenty of other forms of reinforcement learning, I'm sure they have it dialed in.
First we couldn't use our Claude subscription with anything but Claude code, then the limits seemed to change every week without any communication, then they banned a bunch of people (including some prominent names). Then they complain about the Chinese distilling using their API (which I'm partly sympathetic to but let's not pretend that Antrophic invented their training data from scratch).
Then there's this half-baked offer. I mean sure, it looks nice on paper but given how incredibly valuable opensource has been for them and given their budget it does seem a bit tight.
Charitably this will be great for open source software so... so long as they never moat up and lockdown.
A 6-month trial isn't showing appreciation for OSS any more than "first crack hit's free" is showing appreciation for what a good person you are. It's just "you look like a promising customer".
It would be showing greater higher quality appreciation to offer an ongoing benefit.
But there is some benefit to giving maintainers a generous trial length with your offering. 6 months is certainly long enough to see how well it does or does not incorporate into your project.
It just so happens we almost all universally love the offering.
This would be fine in the context of a general sales pitch/marketing deal.
But OSS development and maintenance is special here. It has a budget of $0. As a sales strategy, Anthropic would be better off trying to sell luxury gold plated bindles to hobos.
And there's another question: How exactly does Anthropic see the future of OSS, with this pitch? What are they thinking? Is this the new norm for OSS a $200/month entry fee?
Because adding such a cost to OSS would not only go against everything OSS stands for, and would push the vast majority of maintainers into quitting their projects.
(Now, Anthropic can't mandate maintainers use Claude, though a much-discussed side effect of tools like Claude has been the increased burden on OSS maintainers. And while Anthropic does not raise suggestion that they deal with this by employing AI tools, bystanders most certainly have.)
How many total developers does that cover? 100? How many of them aren't already corporate employees?
And also
> 6 months of free Claude Max 20x
So basically a free trial.
When Github Copilot first launched they gave Pro subscriptions to everyone that regularly committed to a public repo, regardless of the number of stars or downloads, and kept renewing it indefinitely. I don't know if that program is still around but it was amazing to get to try out some early LLM coding tools for open source development.
NPM probably adds a lot. I can't find any recent sources, but NPM packages get downloaded a lot (e.g., every Github Action run.) And to get such a download, an NPM package just has to be somewhere in the dependency tree, which are famously enormous. (Though many might not be updated in the past 3 months, though.)
[0] https://github.com/search?q=stars%3A%3E5000+sort%3Astars&typ...
This is going to get abused so fast, it will make your head spin.
EDIT: I just look up the highest-ranking "buy GitHub stars" page (which I will obviously not link here), and it looks like you would have to pay a little over $1000 to get the required amount of stars. So I suppose it might not get abused as easily as I thought.
On the other hand, someone with the gumption and elbow grease to abuse this process themselves could still easily do so, I'd wager.
All that being said, I still think that GitHub stars are effectively worthless, and attempting to assign value to them like this is, at best, a fool's errand.
I can imagine this will invoke Goodhart's law, increasing the amount of people shilling their AI-generated shovelware onto a Web already greatly suffering from the consequences of the plummeting cost of intelligent-sounding text generation.
They are silent on whether you can prohibit them from training on your input, so I assume you can.
My guess is, if even 10% of maintainers forget to disable training, then Anthropic will have a most excellent source of how really good developers approach problems that can be fed back into the model. That could improve things for everyone.
Of course, the whole purpose of a trial is to induce dependence on the service. Let’s hope that doesn’t reduce the skill of those maintainers. If open source code gets better as a result, that’s good for all.
And of course they're also going to train on your private inputs. It's right there in the TOS.
I previously ignored a free offer when Claude reached out to me as an open source maintainer as it was a glorified free trial. I hope this one continues beyond the listed 6 months, I am not interested in a glorified free trial and if it requires entering credit card details I won't be signing up.
And then Anthropic (and others) comes along, files off the name, and repackages that gift to sell to someone else for money. That’s not respectful.
even if you've got an outstanding project, now everyone has to wade through no much noise it'll never be found
Maybe get out of your SVG bubble and realize that people don't like companies that rat fuck the commons for a quick buck. It's disgusting.
There should be a penny tax per prompt that funds open source development through grants.
pour one out for us gitlab users :(
As a tinnnyy plug, I’ve ran OSS sponsorship programs before for companies. One thing that I always hated was the sales contact process to get it. So, for Vizzly I made it 100% automated. Sign up, connect an OSS public repo, get a free plan. https://vizzly.dev/open-source/ I don’t wanna talk to you and you don’t wanna talk to me (for this :p)
EDIT: Just another test, one of the most used codes in astro -- an ensemble Markov-Chain Monte-Carlo sampler https://github.com/dfm/emcee has 1600 stars. It just shows the 5000 stars is a bit PR, rather than a serious attempt to help open source.
That said, they do have a "contact us" line in there which implies some flexibility.
And completely excludes projects not hosted on Microsoft's GitHub or NPM (Though they do say you can contact them if you don't meet their insane criteria).
Plus, while 6 months is better than 1 month, why isn't it a recurring deal (or token-limited), which renews after check-ins (like educational discounts do). This sounds like an Apple TV+ offer you get for every Apple product you buy. A hook, more than a treat.
In this case, I guess it's just a slimy approach to building a self-selected lead list of people you can hard-hit with upsells after the 6 months.
Thank you for everything you ship*
*there's a 6 months limit we have on gratitute.
https://www.anthropic.com/claude-for-oss-terms
That doesn't mean they're not going to continue this, it just means they're being careful not to make promises which they'll want to roll back later.
What if we get proven code some other way?
Give our tools for free to prove their worth
No one will guess this is astroturf
A special program, with a special account
To get labeled data worth a big amount
No, thanks. I decided I don't want to play those games. I get MiniMax unlimited for 10$ per month, and free GitHub Copilot as an open source maintainer and contributor.
I don't need to beg to get some free stuff, only to later realize the only way to use it is through the shitty Claude Code.
–Ackbar, open source software maintainer
https://github.com/cocaine
Sad day. I hope so they are gonna change the TOS and punish anyone with a 1 million $ fine if someone lies.
That's the only way: criminal charges for students using AI(when forbidden such as academia) and people who plan to abuse it (stealing tokens against TOS).
it's impossible to compete with cheaters and with cheaters who stole money
Sincerely,
Sales & Marketing
In the past week (besides the constant slop), there are models which have misattributed the copyright of new files to me, and stripped my copyright from existing files. It's sapping up time, energy and motivation.
I went for their 100 USD paid tier and it's honestly been immensely useful (Claude Code with the desktop UI with multiple parallel tasks), I've done more and with better quality in the past few weeks than others do in a month - maybe I just got lucky with the domain but it really is a force multiplier and I'm working on like 4 projects in parallel at work and am crushing it, being overworked aside.
Finally I also have enough capacity for various side projects and utility tools/scripts, or at least I will until I burn out, but that's not really the fault of the tool, rather the amount of work.
Being able to throw the latest Opus model at every problem is also really, really nice. Way better than any of the slop before.
It also makes sense to give tools for open source developers. Sometimes we need to test compatibility (does my repo play nice with that harness/ide/etc?). This in turn makes that repo be more solid for the paid tool, which is a potential way of attracting users for both. It has been done by others (like JetBrains IDEs).
There's nothing wrong with taking advantage of limited offers.
In any case, the fine print says that participants have to purchase after the expiration of the free period in order to to continue. Nothing is mentioned about having to give payment info upfront, such that the account automatically transitions to payment.
Participants who are already paying customers will have their payments suspended for that period, so I think for them it will automatically lapse back to paid, at least if their payment method is up-to-date.
Laughable.
This is a tiny, if even unimportant, fraction of the FOSS community that runs the modern tech stack.
Your CEO has bragged multiple times how your tool will make me unemployed. Why would I participate in that?
You stole my code without attribution. Why should I use the services of a copyright infringer?
Sad day. I hope so they are gonna change the TOS and punish anyone with a 1 million $ fine if someone lies.
That's the only way: criminal charges for students using AI(when forbidden such as academia) and people who plan to abuse it (stealing tokens against TOS).
it's impossible to compete with cheaters and with cheaters who stole moneyl