Viewing profile — redox99

I think verifiable tasks like math will soon be like Chess. Machines way beyond human intelligence.

Yeah but open source is often a few guys working on their free time, or a large company open sourcing libraries that they wouldn't be able to sell anyway, or donations from corpora…

> This result is some evidence that AI will not solve RH soon. If there were any easy solution hiding in plain sight then it probably would have found it. There's no way you can co…

There's nothing fast about GPT 4.1. It's ~50 tps AFAIK. Of course it doesn't use reasoning, but you can run modern models without thinking as well. GPT 5.6 Sol without reasoning sh…

Pro: It's web chat only, I haven't tried it so idk. I think it's useful for Math and that kind of stuff, but I only do programming. Ultra: Might be useful, the one time I tried it,…

Yeah, it's hard, there's so many permutations. But most are bad so it narrows it down. Luna: Can be useful if price sensitive, always use with AT LEAST high effort. But codex plans…

It's not just about it "being smart enough". It's about there being actual user demand when it needs to compete with the shiny new model. A 10 year old iPhone is probably good enou…

Yeah I had to double check, o3 feels like it was ages ago. But GPT 5 came out Aug 7, so it's only one day off from my 1 year ago cutoff!

Is there any LLM from exactly one year ago that would be worth running? In Aug 2025 you had - OpenAI o3 - Opus 4.1 - Gemini 2.5 Pro - Grok 4 Even if those were almost free to run, …

Yes. You're better off running Luna Max or Sol Medium and never touching Terra.

So instead of just getting an answer, I have to wait until it asks me, and I have to type back a response? Extremely annoying.

Yes

Terra is awful and not cheap enough to make up for it. I'd suggest using luna or sol.

Because the model can't read your mind and know if you want a quick answer, or an hour long deep dive.

I'm not sure I agree 1) Back then, even as a free user you'd be able to use the strongest model (even if with tight limits). Now, you need to pay to use Sol, and you need to pay to…

Yeah it's definitely better in that regard, but it has the other disadvantages: oversized for single passengers, and also personally it feels like public transport, it's the opposi…

But why include Opus then?

Other than attention optimizations and other minor changes, the top Chinese models (which are way better than gemini) have basically the same architecture as GPT2. Of course RL is …

The cybercab would be almost perfect if it weren't for it's low height. It's easier for the elder to get into and out of taller vehicles.

I watched a bit of it and it's mostly stating their own conjectures as facts. There's nothing objective about it, it's just pushing their own personal agenda.

I think the SOTA is fable max. It's up to you whether that's good enough or not. Almost all other models do make some egregious mistakes with the bike. https://static.simonwillison…

The UI is the equivalent of "You're absolutely right!". It's a big turnoff at least to me. That toast is such an AI thing.

I tried the codex extension many months ago but I didn't like it. I use mostly the Codex App, or sometimes the Codex TUI inside the IDE terminal.

Cursor has two benefits 1) It's still an IDE. Because of pricing I mostly use Codex, but I always have a VSCode/Cursor IDE open, thus have to juggle between the two. Working direct…

They basically don't exist in the currently most profitable LLM market (coding). Yes, subs like codex are heavily subsidized. But API billing has massive margins and that's what en…