I have been hosting gitea for our small org (docker + sqlite) for close to 5 years now. ~50+ repos, some large. All upgrades, including major changes, have been as easy as just changing the tag and reloading the container.
If using outside of subscriptions, It is almost a no brainer to use this simple because of the efficiency. With subscriptions, not so good. Even the officially endorsed OpenAI integration lacks support for server-side context compaction (without a separate plugin which still triggers the pi native compaction).
I added that in https://rcarmo.github.io/projects/piclaw a few weeks back. The reason it’s not in pi-ai is that server-side OpenAI compaction is encrypted/opaque and (at least in my experience) can cause issues when switching models.
I was one of the commenters that was staunchly against this proposal on the original issue that was brought up 3 years ago. Imo their entire argument has always been straw man. Anyone who has spent enough time around the core Codeberg community knows how ideologically motivated their decisions are regardless of what they put out in their PR statements.
This incident coupled with the Forgejo one (In which they twisted the whole story and made Gitea look like the biggest villain in FOSS when in fact they orchestrated the whole drama by submitting a controversial PR that many people don't know about) made me quit and keep away from their hosting.
It's a shame that they exfiled private data. The model is actually good (better than opus 4.8 imo) and the harness itself is butter smooth with the potential of being the best out there.
I know this is what the leaderboards say, but empirically, I'm also seeing it make silly mistakes that opus and fable never do. So it really feels more like sonnet to me as well.
I agree. I subscribe to SuperGrok but never used the grok models a lot for coding. Now with 4.5, I’m gonna hit my weekly limit tomorrow and even considering trying SuperGrok Heavy
I had a very weird experience two days ago where Cursor-Grok-4.5 was either stuck in a loop (it would keep attempting to answer the prompt over and over), or else it would just quit halfway through a reasoning loop. Might have been that I was using omp, but it's still not the most stable thing out there.
Nonetheless when it's working, it's pretty good, and for the price ($10 a month) is an absolute bargain.
How do you know? Presumably you must be able to point to some government transparency documentation about that activity?
Oh wait, nothing like that exists. You don't know what happened. And that's the problem. Neither Musk nor Trump are the CEO of the US. Their behavior should not be considered acceptable by any US citizen.
Please don't sealion. There's no evidence for it occurring in the first place.
It's entirely an internet meme based on extremist headlines. What happened is they set up servers inside government offices that allowed more flexible agile software development environment, then they suddenly got accused of exfiltrating data.
That's what they did to upload a repo, the question is why they would upload the user's entire repo in the first place. The idea itself is egregious, and the implementation was horrifying on top of that.
The entire directory (not repo) and all repo history regardless of secrets or instructions otherwise is not needed for faster cache access. I'm not going to say that this was intentional exfiltration. Sneaky motivations are not needed for this to be a bad idea and a horrific implementation.
reply