From my experience, about 20-30% of the packages are not working with swiftpm, either because they don't have a Package.swift file, or because it is not compatible with up-to-date developer tools. On many projects, I had to fork a few repositories just to add or fix the swiftpm integration... while their Pod integration has always been working well.
- GPT-5 outperformed Opus-4, Sonnet-4, and OpenAI’s O3 across a battery of 300 varying difficulty, error-diverse pull requests.
- GPT-5 scored highest on our comprehensive test and found 254 out of 300 bugs or 85% where other models found between 200 and 207 – 16% to 22% less.
- On our 25 hardest PRs from our evaluation dataset, GPT-5 achieved the highest ever overall pass rate (77.3%), representing a 190% improvement over Sonnet-4, 132% over Opus-4, and 76% over O3.
Why should they have restricted the marketplace? It's really annoying imo that they lock vs codium out of the more useful plugins like the SSH remote one. However luckily most only take a setting or two to enable anyway.
OSS model do not have to be local models, and it's not just about privacy, imo.
DeepSeek R1 hosting is out of reach for most, but it being open is a game changer if you are a building a business that needs the SoTA capabilities of such a large model, not because you will necessarily host it yourself, but because you can't be locked out of using it.
If you build your business on top of OpenAI, and they decide they don't like you, they can shut you down. If you use an open model like R1, you always have the option to self host even if it can be costly, and not be at the mercy of a third party being able to just kill your business by shutting down your access to their service.
Another benefit is they can be fine tuned. Also it's not only about if Openai will shut you down but decide to deprecate model (like they will do for gpt4.0) or swap the name for different model (like sonnet 3.5 did) or censure it or limit capability.
reply