mgambati

It’s apple fault that swift and swift ui is completely garbage.

Not true at all. Do you really believe that generating 1m output tokens on sol really cost 100 usd?

These companies are printing money on inference. The only issue is the capex of expending on new servers but they are generating TONS of income.

Just build it

Free cash? Crazy. I think I can live with the interest of that in the bank.

The inertia is legal and financial. People are paying Anthropic through AWS accounts because the simple reason of not dealing making new contract and legal agreements is enough of reason of the inertia.

But, eventually, I’m quite sure that AWS will also provide open models with those contracts without any inertia. Copilot is already offering Kimi.

My company has a deal with Devin and they provide new models all the time, and open models are becoming the most used ones by our internal metrics, especially because the company is very worried about cost.

Yes! Crazy typo, sorry.

Ultracode is pretty good. I’m not basically using the grill-with-docs from Matt and default plan mode on Cc or codex. It’s just enough.

Not if happens before the new memory factories get built

That’s context, not philosophy.

With 2 wouldn’t have good results. Ideal range for coding is at least Q8.

Nothing weird. The mayor wanted something brag about. That Rio, my friend.

Well, I don’t speed. Had a accident once that made me learn the lesson.

HD riders are known for not using decent safety equipment and that bullshit open helmet or none.

A freaking motorcycle with 300+ kilos moving ate highway speeds or more.

If same on same line, it will block ask to stack branches

Gitbutler still a better option than any worktree like variant

5.3 is good but talks like a robot, it’s too hard to understand what exactly it’s talking about. When using droid I use it to act like worker model and does a great job.

All 5.x models suffer from weirdness in the way it writes but 5.5 and 5.4 are much better and now offer a good balance, direct but without being like Claude.

GPT 5.5 and 5.4 are such great models. I just tried opus 4.8 and took 30 minutes to be confronted with a bit laziness that makes me go crazy. 5.5 just doesn’t have this issue.

They demoed today 8i running ate 1300 to 1600ish tokens per second. I imagine that is caused by having a single rack serving the model just for the demo.

And you still don’t get the quality of data that cursor have which is the best due to being collected pre vibe coding.

Words on the street is that xAI will buy cursor.

The model already advertises because they where trained on massive data’s that refers big brands.

Ask for suggestions for a new pair of shoes. What brand do you think it will suggest Nike, Adidas or some random small one?

I kind feel the same. I’m learning things and doing things in areas that would just skip due to lack of time or fear.

But I’m so much more detached of the code, I don’t feel that ‘deep neural connection’ from actual spending days in locked in a refactor or debugging a really complex issue.

I don’t know how a feel about it.

Funny if you open their website and go to April 2026 you literally see this: 26b revenue (Anthropic beat 30b) + pro human hacking (mythos?).

I don’t think predictions, but they did a great call until now.

It’s funny how you train a machine to mimic human behavior then marketing team decides to promote it “Look! It’s human! Look how it thinking about existence!” while a huge percentage of humanity produced content is exactly about the uncertainty of human existence and that got used to train the model.

Is composer 2 any good? Can it be compared to opus ou gpt 5.4?

They can try add as much as friction they want. A simple rename in the files and directories like .claude makes the thing work to move out of CC.

It’s not like moving from android to iOS.

1m context in OpenAI and Gemini is just marketing. Opus is the only model to provide real usable bug context.

This type of thing is the one I look to see if it’s ai generated. Man didn’t even had the care to type.

If code is getting cheaper, making cuda alternatives and tooling should not be very far. I can’t see nvidia holding the position for much longer.

No politician alive today of any place would understand it