3 ms·
In six month deepseek won't be sota anymore und usage will be wayyyy down.
by Bombthecat 6mo ago
In six month deepseek won't be sota anymore und usage will be wayyyy down.
- man4 6mo ago[dead]
- Barbing 6mo agoWell, if they distilled once…
- Palmik 6mo agoOr there will be DSv4.1/2/3 ;)
- randomgermanguy 6mo agoDefinitely something in this realm, they call the models "preview" at a bunch of different points in the paper. What im really hoping is for a double-punch like with V3 -> R1
- randomgermanguy 6mo agoOnly comparing on SOTA scores (ignoring price etc.) is like choosing your daily-driver by looking at who makes the fastest sports-car...
- dnnddidiej 6mo agoNot really. SOTA vs non SOTA is "can I get my coding work actually done today" vs. "this can do customer support chat" It is like car vs. kick scooter.
- regularfry 6mo agoIt really isn't. We get coding work actually done today on Opus 4.5. That's not SOTA any more, and anything proximate to that level, even quite loosely, is genuinely useful.
- dnnddidiej 6mo agoOK we are in Opus 4.5 is not SOTA. Right by that definition .... yes you are right.
- randomgermanguy 6mo agoI mean its almost halve a year, i think that counts ?
- dnnddidiej 6mo agoTime wise you are correct.
- randomgermanguy 6mo ago> "can I get my coding work actually done today" vs. "this can do customer support chat" I think you need to define "can get coding work done" for this to make sense. Ive been using GPT-3 back-then for basic scripts, does that count ? Or only Claude-Code ? I also think this is a false dichotomy, if you look at the Project Vend project or Vending-Bench, customer support etc. is at no means trivial. (Old but great story https://www.businessinsider.com/car-dealership-chevrolet-chatbot-chatgpt-pranks-chevy-2023-12 https://www.businessinsider.com/car-dealership-chevrolet-cha...)
- UlisesAC4 6mo agoThis, I have been doing my side hustle code with open code an 3.2 reasoner and it is way better than what I have at day job with copilot and whatever models are there.
- dnnddidiej 6mo agoTell me more please!
- wahnfrieden 6mo agoCopilot is a bad harness that perverts the productivity of models like GPT 5.5.
- zrn900 6mo agoNot really. The current SOTAs are already at the point that they can do that. The following models will start to surpass the daily work level. It's a diminishing returns situation just like anything else in tech.
- LinXitoW 6mo agoThe constant improvements of SOTA are the main thing keeping the investment machine running. We can't really remove training costs from inference costs, because a bunch of the funding and loans for the inference hardware only exists because the promises the continuous training (tries to) provides.
- 2ndorderthought 6mo agoA huge proportion of those scores are gamed anyways. Use whatever works for you at the price and availability you can afford