@iruletheworldmo chocolate is impressive, i think we can assume it's grok because it has NO safety guardrails built in.

@KirscheVerstahl Reddit was good like 10 years ago, every large subreddit is ran by the left now though and you can't even post X links on most reddit threads now because they want their safety bubble and are scared of anything but an echo chamber.

@testerlabor source of this? but regardless, Grok 3 is already better than deepseek r1/o1-mini even if o3-mini high+ are better; been testing in lmarena and it answers so much better than other models right now.

@OldDrag64488551 @KirscheVerstahl Sounds like that's a bizarre line of work. Tips need to be taxed just like any other income, ideally tips don't exist at all.

@DSucks32 @Qusaismael @kimmonismus It's in testing right now? It's usable by you if you know where to look and have patience. I'm consistently voting it as top answer when it's being compared vs others like o1/gem 2 pro/etc.

@DSucks32 @Qusaismael @kimmonismus o3 wasn't the competition at the time. Grok 3 is still nearly unlimited use o1/r1 (but better), far better than just being limited to 50 MINI o3 uses a day for $20 a mo.

@tokotoko4000 dang, already got a deep cut with the mystic eyes of death perception, but what about the new waifu from 100gf?

@holguinmora @kimmonismus @elonmusk Then those people are living boring lives restricted by information fed to them by big brother and not thinking outside of the box. I don't care for commoners. I care for AI usable by the people of 4chan & culture.

@Qusaismael @kimmonismus ...no this value is pretty poor when compared to competitors. With X Premium you get 200 requests per 2 hours... I constantly run out of requests on ChatGPT, I never run out on X. Grok 3 FTW (when it's finally out, but lmarena chocolate/kiwi look amazing so far)

you underestimate how censored chatgpt is, you've gotten used to it and know the guidelines at this point. You'd never ask it for anything that leans into nsfw territory or involves any kind of gore/violence, even in a fictional setting, etc. Uncensored AI is a serious need, and Grok 3 surpasses o1/r1 in capabilities anyways & will soon advance open source when they release it with the release of Grok 4, unlike ClosedAI who haven't even released GPT-3 yet afaik.

@johnnythunde247 @ShitpostGate i guess it embeds on phone but not on desktop.

@johnnythunde247 @ShitpostGate You still can though? just paste link with ctrl shift v instead of ctrl v.

@holguinmora @kimmonismus @elonmusk Because I research "forbidden" topics daily as well as I want it to reply to me like a 4chan poster which only Grok can handle.

@SMT_Solvers @dhtikna @crazyshitbro @teortaxesTex @xai not familiar with Mac coding, this good or bad?

@FuriFerntail Well, obviously if they're backed by a corpo they have to keep it secret due to company policy. If not then they might want to have a different vibe for each of their channels or something.

@dhtikna @crazyshitbro @SMT_Solvers @teortaxesTex @xai It's better than they're making it out to be though. coding definitely surpasses r1, maybe not o3, can't test o3 much. Creative capabilities are off the charts too, especially with niche series knowledge being kept in mind.

@NotBrain4brain Meanwhile if it is I am very impressed. It handles creative tasks better than all the other models with top tier coding capabilities.

@testingcatalog @vereskun34734 @lmarena_ai o3/opus are too "ethical". I asked many things including racist/religious things and chocolate/kiwi are only ones that answer without complaining when others give one sentence denial. Very certain this is grok/mistral

@kimmonismus this worldwide like china n all? whatsapp being so popular makes zero sense, I've never heard of any US citizen talk about whatsapp before.

Grok is uncensored, current Grok 2 is better than 4o, full 4o capability but no message you send will ever be rejected for any reason. Grok 3 looks to be very good with the same ideals behind it: NO CENSORSHIP. This is huge. If you try to ask chatGPT/claude/etc illegal or "immoral" questions they will refuse to answer.