tempaccount420

6 ms·

31.
by tempaccount420 6mo ago
You're probably joking, but this is interesting. If we throw more RAM at AI, it can help us optimize programs to reduce our RAM needs, I haven't thought about it like that
32.
by tempaccount420 6mo ago
The article says "POSTED ON MARCH 30, 2026"
33.
by tempaccount420 6mo ago
Benchmarks are not interesting in deciding the "size class". Bigger size means more knowledge. Also, the Qwen 3.5 27B is a dense 27B active parameter model. StepFun 3.5 Flash has 11B active parameters.
34.
by tempaccount420 7mo ago
You're confused, AI can't itself hold copyright, but the human who triggered the AI to write the code holds the copyright instead.
35.
by tempaccount420 7mo ago
You overestimate Wikipedia.
36.
by tempaccount420 7mo ago
Unless the repository uses GitHub's CI. Then it's extremely useful, could be used as RL environment.
37.
by tempaccount420 7mo ago
Politics are indiscriminately flagged. People don't come to HN to read politics.
38.
by tempaccount420 7mo ago
They only need one country for veto rights.
39.
by tempaccount420 7mo ago
Ah yes, a document titled "*THE* Open Source Definition", describing *THEIR* definition of open source.
40.
by tempaccount420 7mo ago
Can you get replacement batteries from Apple? I would not want some no-name brand batteries so close to my ears.
41.
by tempaccount420 7mo ago
What I'd like to know is if there's still a bandwidth issue, where if you use the microphone, the audio quality downgrades severely.
42.
by tempaccount420 7mo ago
How did you type this from the lock screen?
43.
by tempaccount420 7mo ago
Wow, 512GB of storage on the base model! That means the more reasonable 1TB option is cheaper now (+$200 over base).
44.
by tempaccount420 8mo ago
This is not unique to LSPs, a good compiler will do the same.
45.
by tempaccount420 8mo ago
Why does it need to be a daemon? Why isn't the on-disk cache enough? Running a compiler without making changes is instant. Many operations of LSPs feel slower than compiler's incremental compilation. What cache is needed that the
46.
by tempaccount420 8mo ago
LSPs suck. Compilers should be able to re-use their cache for queries, interfaced with through a simple CLI. They should not need more RAM than what the compiler uses. Most LSPs don't work well for big projects or with non-standard set
47.
by tempaccount420 8mo ago
> They do inference at a loss They don't, inference is cheap, especially for agents because of cache hits. The API prices are just inflated.
48.
by tempaccount420 8mo ago
Didn't the safety-conscious employees already leave when OpenAI fired Sam Altman and then re-hired him? In my mind the only people left are those who are there for the stocks.
49.
by tempaccount420 8mo ago
I'm guessing at least 50% of the "users" of Antigravity are actually OpenCode users exploiting the oauth and endpoint. Must be infuriating to them if they're subsidizing it. The OpenCode plugin (8.7k stars btw!) even adv
50.
by tempaccount420 8mo ago
Looks like someone gave their OpenClaw a link to this thread.
51.
by tempaccount420 8mo ago
The disgusting, gross lewdness of a naked body. Thank God someone did something.
52.
by tempaccount420 8mo ago
An employee told him they need to rename ASAP or lawyers will get involved. Is that much different from a C&D? Even if he received a formal C&D, he could have done the exact same thing (rename) and be fine.
53.
by tempaccount420 8mo ago
Well done, you managed to tie Rust, Nix, Haskell and OCaml to "extremely complex, wasteful, and unusable"
54.
by tempaccount420 9mo ago
It's more like the choice between "the" and "a" than "yes" and "no".
55.
by tempaccount420 9mo ago
You would have to pay the API prices, which are many times worse than the subscriptions.
56.
by tempaccount420 9mo ago
They can and already place limits on the server side.
57.
by tempaccount420 10mo ago
Internet is used to defraud people, to propagandize them, to steal their intellectual property and livelihoods, to systematically deny their health insurance claims, to dangerously misinform them (e.g. illegitimate legal advice or hallucina
58.
by tempaccount420 10mo ago
It's not in Google's style, but they need a codex-like fine-tune. I don't think they have ever released fine-tunes like that though. The model is very hard to work with as is.
59.
by tempaccount420 10mo ago
They had to rush it out, I'm sure the internal safety folks are not happy about it.
60.
by tempaccount420 10mo ago
Coding requires a mindset shift that the -codex fine-tunes provide. Codex will do all kinds of weird stuff like poking in your ~/.cargo ~/go etc. to find docs and trying out code in isolation, these things definitely improve capab

More ›