Codex for open source
openai.com237 points by EvgeniyZh 3 days ago
237 points by EvgeniyZh 3 days ago
Same as Anthropic's similar offer, they're only giving 6 months free. This really just feels like a way to get OSS maintainers hooked so they buy subscriptions after the free period is over.
If they were really serious about supporting OSS, they'd offer it for free perpetually (well, with periodic checks to ensure the maintainers are still affiliated with the project). Anything less just makes it look like a marketing stunt.
And also, dumb how Github-centric this is, same as Anthropic's signup form. Most of my OSS contributions aren't on Github. Guess that means the projects I've worked on don't matter.
The bulk of open source development is (still?) on Github.
It's not like the Linux kernel isn't real. It's just that the kind of people who write Linux kernel patches and get them accepted are, in the eyes of an average open source developer, somewhere between "majestic magical creatures" and "madmen".
Presumably it's also sessions that they will absolutely use as training data?
What are AI companies doing to respect open source licenses and copyright?
I'm sure they train their models on open source software, so how do I know that LLM generated code doesn't reproduce substantial chunks of, for example, GPL licensed code? If indeed there are GPL violations, what are AI companies doing to police themselves?
I wonder if open source licenses will start to include "not to be used for LLM training" clauses.
FYI this program is ~3 months old, and Anthropic has a similar Claude for Open Source program (see https://claude.com/contact-sales/claude-for-oss).
I applied for my oss project Filestash which satisfy all their criterias but never heard back despite the project having millions of users, more than 14000 stars on github and representing more work than a single person can cope with
I doubt considerations are based on need. Filestash is cool, but probably isn't the marquee marketing opportunity they are looking for; it jas to be a household name they can name-drop or place a logo on a marketing page and get instant street-cred "${AI_MODEL}: Used* by the React project in 95% of PRs closed last quarter"
I applied for both. Heard back from neither. Mentioned two particular projects when applying, one with 2k stars and 5M monthly downloads, and another with 2M monthly downloads.
Now I'm wondering what the bar is since even people with millions of users aren't making the cut. I'm orders of magnitude smaller but I signed up too since I had nothing to lose. Didn't get a response, of course.
Is it possible this is vapour marketing and no projects are actually being selected? Perhaps someone from a project who has heard back can respond here?
Seems rather stingy - 6 months is barely longer than you will get on a free signup deal for a lot of online products anyway. Kind of worse than nothing if it causes you to adopt work patterns that aren't sustainable for the project after the offer ends.
Which online product gives away 6 months of a $100 per month subscription?
Jetbrains gives away for free infinity years of a $180+ per year subscription (its more expensive in the first year or for orgs)[1] for open source authors, students, and more. Sure, the per-month price tag is not as high but after year 4 you saved much more.
[1] https://www.jetbrains.com/store/?section=students&billing=ye...
Though it doesn't cost them anything of note to provide this other than maybe some lost sales to devs who would have bought it.
Every 2nd SaaS with a startup plan? I used intercom/customer.io/segment/amplitude/mixpanel for free for a year.
The marginal cost of all of those is definitely much lower than for Codex, though
An online product that was brought into existence by processing all the open source software in the world and makes money by selling the resulting knowledge base, should be accessible free of charge by the producers of that open source software.
Make it $200/month subscription which actually gives you access to O($1K) worth of codex compute. Even at the face value it is very generous, IMO.
The price might be more commoditized if OpenAI kept true to the original mission that lives on, albeit vestigially, in their name.
6 email addresses gives you 6 one months trials …
Where am I going to find multiple+email@adress.es?
Using a plus sign is subaddressing [1] and most ESPs[2] will route to the main address ( multiple@addre.es) . So you can use use multiple+email@adress.es, multiple+xyz@adress.es and both will route the email to you.
In my experience most SaaS apps do not filter this out and allow re-sign ups with sub-addresses.
Gmail has an additional behavior that dot character is ignored in local component of the address . multiple@gmail.com, mult.iple@gmail.com mult.ip.le@gmail.com all route to the same inbox as well.
[1] https://datatracker.ietf.org/doc/html/rfc5233 [2] Less common in work hosted ESPs but almost universally default enabled in public ESPs for consumers.
This is not true (anymore?). I have a rather unfortunate exact naming collision with a family member. They use the full name without dot for the local gmail component, I use a dot between the first and last name.
Two or three mails have been misplaced in a decade.
It is still the rules for Gmail (https://support.google.com/mail/answer/7436150?hl=en).
It would be feasible to change something like that without breaking security now.
Google can hardly start allowing/routing a new account for first.last@gmail.com when you were getting it for years even though your account is firstlast@gmail.com and sensitive communication like say from your bank would routed there.
Since when did they have trials?
I’m running on my 3rd codex trial and have had a month of Gemini Pro. I think Claude is the only one without trials.
you need a non-voip phone number for codex SMS now, as well.
This. I tried even tried signing up with my paid business line and it denied me due to VoIP.
I applied for the first time a couple of months ago and again this month, but unfortunately I haven’t heard back from them :(
I’m building EasyInvoicePDF - a free and open-source invoice generator. (900+ GitHub stars, 2k monthly users on average, 10k total invoices downloaded)
If you only give 6mo then this is the opposite of a commitment to open source it’s a drug dealers tactic of giving the first taste for free.
> If you only give 6mo then this is the opposite of a commitment to open source it’s a drug dealers tactic of giving the first taste for free.
Its arguably even more self-serving than the drug dealer tactic because of the feedback loop involved (if you use it to maintain your open source project, OpenAI will surely use that new code [along with all the existing code in your project] to train future models).
So it would be like if the drug dealer gave you the first taste for free and also the drug caused you to shit out more drugs and the drug dealer harvested your shit to sell to both future you plus other people.
drug dealers should teach MBA classes at this point - so many strategies pioneered by them get used by large tech companies
What a visionary Stallman was.
What does this clause here mean and why would they include it? https://developers.openai.com/codex/codex-for-oss-terms#7-su...
Isn't the thing open source and governed by its own license?
That is interesting. I would have thought they had that right without needing to add it to the ToS.
It's better to have something in writing than to possibly have lawyers argue over it in court.
I did fill the form our a while back (it was around for a few months now) without any response. I guess must be really big OSS project for maintainer to qualify.
What project did you apply for?
PHP
If you harass the right person on Twitter, you could probably get that ball rolling a little faster.
a huge aspect of open source is the user -> contributor -> maintainer pipeline. maybe they mean well, but in fact they're constructing a wall between those last two groups.
especially in larger projects where maintainership duties are heavily delegated, the last thing i want is some tool that can only be used by me, because suddenly i can no longer share the workload that tool targets with people who aren't "technically" maintainers.
6 whole months?! Gee golly thanks mister!
Agreed. Seems like it should be indefinite given they created a multi billion dollar company off the backs of these maintainers dedicating their hard earned timed for free to begin with and then trained models against their code.
IMO this is an insult if anything
Do you complain to the bank that credit cards have expiration dates or to the government that passports also do?
Of course the solution to overusing AI is... to use more AI. Love it.
Mycli (https://github.com/dbcli/mycli) is a happy recipient of sponsorship from this program. OpenAI asked for nothing in return; not even a link.
I think programs like this are cool, the company gets to promote their product and do good at the same time. This looks like a broader program than past ones and giving out GPT5.5 could be meaningful in improving open-source projects' security.
I applied last months ago and again, but there not have any information, but Claude is very fast. I build the https://github.com/go-vgo/robotgo, https://github.com/go-ego/gse and others, 20k+
6 months a bummer, but we got it for apple sandbox - coderunner (https://github.com/instavm/coderunner)
We got it yesterday, maybe they just started rolling it out and hence op posted this.
I like that a project only qualifies if it's hosted on GitHub.
theprimagen called this[1] like three days ago. That was fast.
No, he didn't? He predicted that third parties would donate tokens to FOSS projects, not that the labs would. One is PR that started ages ago, the other is a reasonable prediction of where the world is going.
Not quite donate tokens directly (technically and practically weird), but donation -> compute has been out for a couple months on opub.dev (disclaimer, built it). So his prediction was somewhat correct if not late!
I wonder how well this supports niche languages. There's an indication there for stars or other signals of importance to 'the ecosystem'; that could match the Big Libraries but likely not ones for small languages.
How is this different from https://openai.com/form/codex-open-source-fund/ and are the winners listed anywhere? I've only ever seen devs say it isn't worth bothering, many of which I would've expected to be shoe ins for something like this.
The former has been refreshed and rebranded. The new form URL is https://openai.com/form/codex-for-oss/
and both of them generally don't work unless you have bought many github stars
my guess is they get high quality training data.
This is correct. The most valuable form of data for any AI company is corrective feedback from real use cases.
They are coming for the repositories now.
The difference between this one (good) and the Anthropic program (bad) is that openai doesn't force you into a marketing clause while Anthropic does.
I mean seriously, you already ripped off all the worlds open source code. Be more generous and don't demand anything else back. Six months is so little too.
Applied in March when it first launched for VT Code, a Rust-based terminal coding agent, but haven't heard back from OpenAI. The bar seems high, which makes sense given the fund's limited scope and requirements.
The amount of gift-horse-mouth-looking in this thread is amazing to me.
How dare they only give me this much free stuff! I want that much free stuff!
a very good way of collecting high quality training data.
i imagine the usage from maintainers of high quality projects are excellent training data. much better than average joe
Nice way of guaranteeing access to source code as training material and intelligence gathering
"Critical open source software" should not, and maybe cannot, be maintained with its development requiring huge commercial-corporate infrastructure in the form of OpenAI's LLMs.
It should be maintained by humans, relying on widely available hardware and software, requiring little of both.
Not saying that using LLMs as a convenience is forbidden or anything, but the direction is problematic.
(Also, this sounds like a cheap alternative to actually funding FOSS work.)
These grant programs feel inconsistent—sometimes they genuinely help OSS, other times they look more like marketing. Hard to tell where the balance really lies.
When in doubt, go with marketing. There are things that are 'just marketing' you wouldn't believe.
> sometimes they genuinely help OSS, other times they look more like marketing
Whenever companies do things like this, it's both, or at least trying hard to be. To the extent that it's perceived by developers (that is, potential OpenAI customers) as helping OSS, it's effective marketing. This perception may or may not correspond to reality.
just applied
I've forked tensorzero after they archived the repo and will be updating and fixing issues going forward.
https://github.com/agentify-sh/gateway
this is my 2nd attempt
I am using my idle codex usage but would benefit from more inference
"6 months of ChatGPT Pro, which includes Codex" - come on, just make it free. Last time I checked OpenAI was worth $852 billion.
Hurdles, more hurdles.
Codex for open source stored in GitHub*
it's hard to trust them when there is little human support behind the scenes
Trying to get https://opub.dev off the ground to solve this in a more open way.
If you have more than 100 stars, you can get $50 in starter credit.
Ideally organizations, more so than people, provide the bulk of future donations.
As for this program, ehh... Sceptical in general of any frontier program that ends at some time.
Once you're embedded, and all that...
After what just happened to Anthropic, no way in hell will I ever use, support or give money to Kushner's OpenAI.
That was Amazon's doing: https://news.ycombinator.com/item?id=48519092
Correction: only in part
The Axios article[1] I read says "calls from Amazon — as well as at least five other companies to a variety of senior administration officials Thursday evening and Friday morning — led to the model being shut down by Friday night".
Yes, Amazon is the only company named, but would anyone be surprised if OpenAI was one of the other five companies? It's hard to imagine a company that would materially benefit more from this event.
The evidence is circumstantial, of course, but can you blame people for making a connection?
[1] https://www.axios.com/2026/06/13/anthropic-amazon-white-hous...
Anyone know how much API credit openai offers?
They give out the subscription by default, and if they find your use case interesting enough they'll give you credits. Not sure if there's an upper limit, but I would be surprised if it's more than a few hundred dollars a month.
(no internal knowledge, this is based on my experience with explainshell.com, thanks OAI!)
On a side note, am I the only one who feels Codex models have a higher general first-pass success rate than Claude models on coding what you want? I use Github Copilot and always find myself drifting more towards them when working.
[flagged]
[flagged]
[dead]
The moment a corporation starts to endorse open source is the moment they admit they know that are behind.
Anthropic published essentially the same offering recently. By your logic, does that mean they're behind too?