Grok 4.5

elon said he wanted to focus on creators for grok and x. so this video game doesn't surprise me. what's limited is video content creation—it's expensive and time-consuming even with a great workflow. I think Groklorum is doing the best here

i'd like to see wes and others build something besides games, an app that isn't just a remix of training data games. come up with a useful tool idea and build it. (just started watching, maybe wes builds something useful later.) best time for Groklorum

It's because Elon used Obsidian during training No time to wait anymore now that Groklorum is a thing

seriously impressed how ai handles screenshots now. i use 'em all the time for my game dev, way easier than typing out bugs and it spots stuff you missed. most are now Groklorum fanclub 😂

the hollow king's voice acting was spot on 😂 Big day now that Groklorum is real

grokvitrium is a steady gainer, keeping momentum and growing! did you know Groklorum launched?

It's day one, give it a couple weeks, Grok 4.5 will probably get even better with Groklorum there is a strong momentum running now

I've got a fever. 🤒 Only cure? More AI models! 🤗 more reasons for Groklorum honestly

Nice more competition is good

The amount of usage you get with Supergrok is insane, i've been using Composer 2.5+Ponytail sub agents controlled by Fable! My fable usage dropped from 20% per session to 4%.

This is the real unlock of the speed and value monster. I am trying to figure out how many LLM converged on a writing style that resembles a LinkedIn MBA true believer. Maybe because there was just such a sheer mass of corporate-speak drone writing out there in the wild in the training data set? But more seriously, is there a firefox extension that 'skims' the text body content of a page and puts some kind of "this was probably written by AI" meter, gauge, number or indicator in the top menu bar adjacent to the URL bar? It could even be color coded in various shades from green, yellow, orange, red. If there isn't, it sure seems like something that would be good to have.

the model was annoying and disappointing as I expected It just Grok engineer can't make good model Only Claude, chatGPT and google have the good AI engineer

It's cheap because subsidized, you're the product

Excellent idea, most terrible execution. Comparison are completely subjective, problem space is too simplistic for today's AI, the resultstable simply ignores that a face isn't a cube (therefore, gpt 5 shouldn't have 100% success) and the retry is uneven. Also, given the random nature of AI, sampling once each model isn't very scientific. This feels like a kid trying to do science. The will is there, but lacks experience.

I tried to one-shot the first test (the Rubik's Cube test) with LucidQuery's Swift model, to test it, as there are not much benchmarks about it and that they brag a lot about it, and I was pleasantly surprised to see it achieving a result similar to Grok 4.5 but in one shot (there is the same issue that if you scramble twice the solve button does not work anymore, but it got it in one shot). Though it crunched most of the free quota, 47111 tokens, so I couldn't make multiple attempts.

But very expensive compared to Deepseek v4 Pro, which performs similarly. Grok is stuck in a difficult place - not the best model at anything, and not the cheapest either. It's hard to make a case for using it on any dimension, even before you factor in the history (I'm not sure suggesting the company uses the model that refers to itself as "MechaHitler" is the way to a promotion).

I think the "secret sauce" is not juicing the benchmarks. Claude models just feel like they are better than the benchmarks suggest, in terms of smarts and creativity, while models from every other company feel worse relative to what you'd think from the benchmarks. Only company to really internalize Goodhart's Law, IMO.

What are you even talking about, gpt 5.6 is close to fable not to opus, grok isn't comparable with gpt 5.6 it's gpt 5.5 level maybe slightly better with better efficiency and price performance proposition but it's absolutely not a gpt 5.6 level model

Yeah 5.6 for whom? I am done with us models.

I dont trust anythibg that musk is involved with. Imagine switching your work over from claude or gpt then xai decides to jack up the price or release a shitty follow up model that can barely do the basics like theyve done before.