Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Grok Build 0.1: Intelligence, Performance and Price Analysis

artificialanalysis.ai

Post reply on HN
Print view View original

11–19 of 19 posts

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#11

Post by vessenes »

khurs wrote: post #5
sourcegrift wrote: post #3

As much as I hate to say this, I'm afraid that Musk is abusing his capital power and Grok Build will end up as the top model eventually. I tried it on a couple of projects and was (pleasantly?) surprised how good it was. I have no doubt in my mind that spacex needs to be broken up before they grow too big because grok build is going to be the future coding model of choice

Musk's 'capital power' is spread thinly, as both xAI and SpaceX need ongoing enormous funding and have debts. Google raised the same amount as SpaceX IPO at the same time, it just didn't get as much attention outside of financial and tech press https://techcrunch.com/2026/06/03/alphabets-record-breaking-... The other big players (MS/Goog/Meta) have solid businesses already and both OpenAi and Anthropic are IPO'ing so…

SpaceX is cashflow positive.

Buying compute that others pay a premium for is not over-buying. Note that $85b GOOG raise is for … building compute they don’t have.

The SX bond issuance reportedly has $90b in demand; they took $25b.

With all of those in mind, it might be worth updating your take on the relative capital power. I’d term it significant. Knowing Musk he will be investing aggressively where he thinks matters over the next few years.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#12

Post by vessenes »

I don’t know the methodology at this site, but I noted with interest they rate grok build slightly below Meta’s offering(!)

I would anticipate we will see aggressive improvements / new models at least as the Cursor integration comes together, and a hard push on price and quality over the next few months. Elon is definitely eyeing revenues from the coding LLM market, and has a lot of structural power, since the competitors are running on his hardware right now.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#13

Post by khurs »

user43928 wrote: post #10
khurs wrote: post #5

Earlier quoted context omitted.

Musk's 'capital power' is spread thinly, as both xAI and SpaceX need ongoing enormous funding and have debts. Google raised the same amount as SpaceX IPO at the same time, it just didn't get as much attention outside of financial and tech press https://techcrunch.com/2026/06/03/alphabets-record-breaking-... The other big players (MS/Goog/Meta) have solid businesses already and both OpenAi and Anthropic are IPO'ing so…

I cannot comment on the likelihood of xAI being competitive as an AI lab in the future. However, their "overbuying" of compute means they can now rent it out for $2.32B/month. That seems like solid business to me, and raises the question whether your claim of xAI needing 'ongoing enormous funding' is accurate.

Most of that is from Anthropic which is IPO'ing?

Prior to IPO they want to show the best growth numbers possible, so how long they continue with the deal post IPO remains to be seen.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#14

Post by blenklo »

vessenes wrote: post #11
khurs wrote: post #5

Earlier quoted context omitted.

Musk's 'capital power' is spread thinly, as both xAI and SpaceX need ongoing enormous funding and have debts. Google raised the same amount as SpaceX IPO at the same time, it just didn't get as much attention outside of financial and tech press https://techcrunch.com/2026/06/03/alphabets-record-breaking-... The other big players (MS/Goog/Meta) have solid businesses already and both OpenAi and Anthropic are IPO'ing so…

SpaceX is cashflow positive. Buying compute that others pay a premium for is not over-buying. Note that $85b GOOG raise is for … building compute they don’t have. The SX bond issuance reportedly has $90b in demand; they took $25b. With all of those in mind, it might be worth updating your take on the relative capital power. I’d term it significant. Knowing Musk he will be investing aggressively where he thinks matter…

Space-X invests a lot into R&D for Starship which they need to make starlink scale.

They also need it to at least show that in theory they can put a datacenter into space (which is stupid because it would take 300 starship flights to bring up their small colossus dc into space which is only 200MwH).

They also increase the price of Starlink, have to shell out money for cursor.

They overbuyed because it clearly shows that Grok is not frontier or people don't like to use it for other reasons.

Of course Musk has to invest, if what he thinks matters and is susccessful is another story like Cybertruck or his Datacenter idea or his dysonsphere 'idea'.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#15

Post by blenklo »

user43928 wrote: post #10
khurs wrote: post #5

Earlier quoted context omitted.

Musk's 'capital power' is spread thinly, as both xAI and SpaceX need ongoing enormous funding and have debts. Google raised the same amount as SpaceX IPO at the same time, it just didn't get as much attention outside of financial and tech press https://techcrunch.com/2026/06/03/alphabets-record-breaking-... The other big players (MS/Goog/Meta) have solid businesses already and both OpenAi and Anthropic are IPO'ing so…

I cannot comment on the likelihood of xAI being competitive as an AI lab in the future. However, their "overbuying" of compute means they can now rent it out for $2.32B/month. That seems like solid business to me, and raises the question whether your claim of xAI needing 'ongoing enormous funding' is accurate.

It clearly shows that the billions investedin xAI are not returning what they should with the product they invested in.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#16

Post by khurs »

vessenes wrote: post #11
khurs wrote: post #5

Earlier quoted context omitted.

Musk's 'capital power' is spread thinly, as both xAI and SpaceX need ongoing enormous funding and have debts. Google raised the same amount as SpaceX IPO at the same time, it just didn't get as much attention outside of financial and tech press https://techcrunch.com/2026/06/03/alphabets-record-breaking-... The other big players (MS/Goog/Meta) have solid businesses already and both OpenAi and Anthropic are IPO'ing so…

SpaceX is cashflow positive. Buying compute that others pay a premium for is not over-buying. Note that $85b GOOG raise is for … building compute they don’t have. The SX bond issuance reportedly has $90b in demand; they took $25b. With all of those in mind, it might be worth updating your take on the relative capital power. I’d term it significant. Knowing Musk he will be investing aggressively where he thinks matter…

Musk didn't become the richest man in the world by accident, he is a formidable talent with deep connections.

But no indication that XAI will lead the field any time soon.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#17

Post by recsv-heredoc »

The harness is really important. It matters so much - possibly even more than the model. We had harness crashes after running many agents - granted we were doing quite a bit with it. Grok Build (as a product) review here:

UX friction behind is worse than Claude Code - but seems to be a strange positioning choice - they're more on the 'vibe' side than the 'agentic engineering' things.

Largest issue was actually reviewing output - but if you're going to largely make that opaque from the user, why choose a CLI-based interface that's so mouse-heavy?

There's also problems with the actual model. Thinking is visible, and every interaction goes like this:

"I would like you to investigate adding an API route to tackle x,y,z" *Grok, thinking: Okay - the user has asked me to add an API route to tackle x,y,z"

Also absolutely absurd other quirks - "I have no tools available in my context" being visible in the CoT.

The auto-approval (yellow, auto-mode) review of Claude Code via Opus is a killer feature - every build-it CLI should be offering this for long horizon tasks.

Messaged one of the engineers about our experience - no feedback.

You'd be better off with Claude Code 5x Max than the 300 USD/month subscription.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#18

Post by user43928 »

khurs wrote: post #13
user43928 wrote: post #10

Earlier quoted context omitted.

I cannot comment on the likelihood of xAI being competitive as an AI lab in the future. However, their "overbuying" of compute means they can now rent it out for $2.32B/month. That seems like solid business to me, and raises the question whether your claim of xAI needing 'ongoing enormous funding' is accurate.

Most of that is from Anthropic which is IPO'ing? Prior to IPO they want to show the best growth numbers possible, so how long they continue with the deal post IPO remains to be seen.

54% is Anthropic.

Considering their models had outages with the message 'model overloaded' just a few days ago, I doubt that their compute demand is ending soon.

Re: Grok Build 0.1: Intelligence, Performance and Price Analysis

#19

Post by grim_io »

Since their very brief rise to the top with grok 3, they have been firing their talent and having their execs leave.

The result being that their models are falling behind further every day. Now they are behind Chinese open weight models.

The allegedly most valuable company can't even utilize their earthly data centers, so they have to rent them out to people who can.

Embarrassing.

Post reply on HN