deaux

20 ms·

451.
by deaux 6mo ago
> Not a single pro-Gaza or anti-Israel activist will acknowledge this Go easy on the Kool-Aid. It's the opposite; those things are not talked about because they are universally acknowledged by anyone except the groups themselves as
452.
by deaux 6mo ago
> but I've found that Claude, at least You hit on a very important point here. The linked AGENTS.md is a bad idea for general purpose use because the things it's meant to tackle, including an inherent bias towards or agains
453.
by deaux 6mo ago
> Sort of feel sorry for Figma in a way though, given all the "partnerships" (highlighting their MCPs) and case studies they've done with Anthropic and then they release this. I note there isn't a testimonial from the
454.
by deaux 6mo ago
Nope. Source: Multiple successful prodiction applications with >90% LLM generated code already in Late 2024/Early 2025. Not slop, not note taking apps. Nothing ground-breaking but real-world valuable usecases. Both by myself and by
455.
by deaux 6mo ago
Sir, this is ~a Wendy's~ talking about paid use for agentic usecases, especially as individuals for work or as small-medium sized companies. Not about people asking chat their horosocope for the next week. Yes OpenAI still has the horo
456.
by deaux 6mo ago
Reddit and X are on the stores. I guess browsers are on the stores, at least on Android where they aren't necessarily Safari reskins.
457.
by deaux 6mo ago
You don't have a growing belief, you have an accurate observation. This comes up often when bad actors promote the meme "everything is securities fraud". In reality, all cases that they're talking about are instances of
458.
by deaux 6mo ago
> and debugging is a much more valuable skill than picking the right tool for the job. Can't agree with that. Debugging is short-term, picking the right tool is long-term. Unless you thought I meant agentic tool ;)
459.
by deaux 6mo ago
The open weights models absolutely compete with flagship labs for most tasks. OpenAI and Anthropic's "cheap tier" models are completely uncompetitive with them for "quality / $" and it's not close. Google
460.
by deaux 6mo ago
Yes it is and has been for a very long time, it has been years now. Gemini 1.5 Pro is when LLM translations started significantly outperforming non-LLM machine translation, and that came out over 2 years ago. Ever since then Google models h
461.
by deaux 6mo ago
I find Gemma 4 26B A4B better than Haiku 4.5 and that's smaller than this one.
462.
by deaux 6mo ago
> If you just do a tiny amount of tok/day and can wait for the answer to be computed overnight or so But they can't? The usage pattern is the polar opposite. Most people running these models locally just ask a few questions to
463.
by deaux 6mo ago
> that's their answer and it seems people have caught on. There's nothing to catch on to. OpenAI have been shouting "come to us!! We are 10x cheaper than Anthropic, you can use any harness" and people don't come
464.
by deaux 6mo ago
You use Claude Code? Then harness changes will have had much more impact than any model "stealth nerfing".
465.
by deaux 6mo ago
Yes, GPT 5.4 is better at finding bugs in traditional code. This has been easy to verify since its release. Its also worse at everything else, in particular using anything recent, or not overengineering. Opus is much better at picking the r
466.
by deaux 6mo ago
> It seems to be a rule that older models are more expensive than newer ones. It isn't. Gemini has gotten more expensive with each release. Anthropic has stayed pretty similar over time, no? When is the last time OpenAI dropped API
467.
by deaux 6mo ago
While they can be run locally, and most of the discussion on HN about that, I bet that if you look at total tok/day local usage is a tiny amount compared to total cloud inference even for these models. Most people who do use them local
468.
by deaux 6mo ago
> but it fails at million-user scale where even that level of human support is too expensive. In general this is a myth promoted by platforms with millions of users. The vast majority of such large platforms could easily afford that leve
469.
by deaux 6mo ago
This can be done with Claude Code just fine. Or simply API usage.
470.
by deaux 7mo ago
The Guardian here using the word "climate" 6 times. They mention prices and economics zero times. They talk about energy sovereignty zero times. While the US attack on Iran has massively increased fossil fuel prices. Framing solar
471.
by deaux 7mo ago
Haha, the Guardian is just as much in bed with the capital class.
472.
by deaux 7mo ago
> Give me an onshore wind farm over this. Guess why those aren't common? Largely because the same people vehemently opposing these solar parks, have already been blocking onshore (and even near-coast offshore) wind for more than a d
473.
by deaux 7mo ago
Would've been nice if you'd read them as not a single one of them even mentions Agent SDK or claude -p usage, the topic of this thread.
474.
by deaux 7mo ago
Really? I'd love to know other good examples, because he's the only one I know who has been so public about it including taking several meaningful actions as a result.
475.
by deaux 7mo ago
I've got slideshows and interactive calculators and things and it all works fine without the islands, that's the thing. LLMs will just incorrectly pattern match "interactivity = React".
476.
by deaux 7mo ago
That's interesting - I thought apps that didn't go through the App Store but were in "developer mode" were time-limited, but maybe I confused it with "share with family" rather than pure personal solo use. So
477.
by deaux 7mo ago
I respectfully disagree. Until very recently they never had a model that came close to Pareto competitiveness at their price points. Recently theh released Mistral Small 4 and for the first time it's competitive.
478.
by deaux 7mo ago
Z.ai, Moonshot, Deepseek - which of these spent trillions? Their best models are much stronger than Mistral's, the first two have models that are competitive with the US frontier ones. Not hating on Mistral here, their latest Mistral S
479.
by deaux 7mo ago
> There are places with the same entrepreneurial drive that are part of the EU. Name them rather than being vague, clearly people here want to hear where they are.
480.
by deaux 7mo ago
This seems pretty limited to tech. For example in domains like media and marketing - I mean startups in these fields, not "marketing team of a tech startup" which is tech - the amount of "drive" feels the same whether th

More ›