Cursor, since Grok 4.5, has had an incredible deal for frontier level models, their subscription now goes way further than OpenAI or Anthropic. Even on their lower tier plans you can use a lot tokens on their of their first party models (Grok and Composer) and not really run out comparatively. Combine them with an orchestrator and implementor type setup and it goes even further.
I believe they are the only western provider that has Kimi K3 on a subscription plan today as well. I would love to ditch Anthropic and be on Kimi if there were a subsidized plan like that with ZDR
Can you explain what you mean? These days courtesy of an addictive reset game OpenAI is playing, I can't find anything with frontier intelligence that's more cost efficient...
I'm willing to pay 2x for a 10% smarter model. Intelligence matters that much (because 10% smarter probably saves, on average, several hours of human time).
For building a full stack custom CRM and media pipeline tool with video conversion, transcription, and indexing. Supabase, AWS, Meili, NextJS, GCS - lots of surfaces and planes.
4.8 basically couldn't do it, I abandoned the project as the fallback was, "current business processes".
With F5 it's been 4 weeks and almost ready for production release.
SpaceXAI is the only frontier model company that had its own compute/date centres and soon chip making factory, I think they will pull ahead with cheaper tokens similar intelligence and better harness/tools. Grok build is 2-5x faster than Claude Code in my opinion.
I have. He was using it due to philosophical reasons the same way many people have philosophical reasons for avoiding it. I don't know how many people are like that, but it's not exactly where you want to position your product if you're a business.
Personally - and I know I'm not alone with this sentiment based on comments I see on this site - I wouldn't touch Grok no matter how good or cheap it is. I don't trust Elon and I don't want to give another dollar to the world's richest person who turns around and uses the money to interfere with elections. The guy I know uses it for essentially the same reason I won't use it.
> the world's richest person who turns around and uses the money to interfere with elections.
I'm no Elon fanboy but to be fair a lot of the richest people in the world heavily fund and endorse political candidates, parties, PACs, lobbyists, etc across the political spectrum. Elon was just far more public about it and it was covered much more.
This statement doesn't go far enough given Elon's direct and hands-on involvement with DOGE and the 2024 elections. Very few of the richest people of the world are personally entangled in meddling with government agencies directly, for example.
> MADISON, Wis. (AP) — Billionaire Elon Musk likely broke Wisconsin law when he promised to hand out $1 million checks to voters in the 2025 state Supreme Court election, a bipartisan panel has found.
> The Wisconsin Elections Commission last week referred two complaints to the Brown County district attorney’s office, which can choose to bring criminal charges over violating the state law against election bribery. Prosecutors have 40 days to report back to the commission.
Versus Sam,Dario or the CCP? Im all for running local models but im sor far from being able to pay for a large model hardware setup. My strix halo box is like driving a beaten up vespa when the frontier models are Ferraris.
There's a big difference between quiet donations to a PAC - not that that's good either - and what Elon did. He literally paid for votes, likely in violation of the law. He poured more money into U.S. elections than anyone has before. He and Trump both made strange, cryptic statements about Elon's role in Pennsylvania with the voting machines that has caused people to reasonably wonder if they somehow manipulated the election. Whether he did or not, the innuendo alone is not ok. Then he did what he did in Germany. Don't even get me started on DOGE or his Starlink shenanigans in Ukraine.
That's before we even start talking about the models themselves. He claims to want "unbiased" models, but he very clearly has a distorted view of the world and has repeatedly demonstrated a desire and willingness to bend the world to his will. I don't want to use a model that is so obviously suspect. Not to mention, his models repeatedly produce racist, Nazi-like propaganda.
IMO, he is, at best, a clueless amateur masquerading as an expert and running into problems a more careful person manages to mostly avoid. At worst... well, you get the picture.
"Everyone's doing it" presented with no evidence is just you trying to make yourself feel better for:
- using from
- driving a Tesla
- voting for Trump
It's a really, really easy line to draw in the sand: don't support openly corrupt individuals.
It's absolutely true there's money in politics. To call all money in politics equally corrupt because Bernie got a dollar to have dinner with someone vs Elon effectively directly buying votes....
A complete lack of nuance here. And I wouldn't be surprised if corporations / super rich WANT you to think like that. The more defeatist the mentality becomes the more we just accept whatever they do next.
A bunch of SWEs at my work use it as their primary model.
We have Claude, ChatGPT, and Cursor with essentially no cap on spend (top guy is spending over 10K a month on AI at API prices), and he hasn't had his hand slapped.
So it's not like they are using it purely because it's cheaper.
I think people like to use it for its speaking style, pretty solid performance, and its speed.
It's much faster, so if you're not doing something cutting-edge, or you're doing the planning yourself and just using the LLM for implementation, the speed benefit outweighs the extra smarts of Fable/Opus5/Sol.
I had a security incident the other day and Grok was the only model that would help. Claude and GPT refused on ethical grounds and only gave general advice. In an emergency, I'd only trust Grok. However, that's the only time I used Grok for coding (since Opus 4.8 it would take a lot to get me to switch away from Anthropic)
I use. I used to be a Claude user. Since trying Grok 4.5 and especially Grok 4.6, I don't want to go back to Claude any more (I have early access to 4.6).
Grok is 3x+ faster than Claude and I can't tell the diff in engineering work quality. As an engineer, speed is important to me.
An hour in, I've been running four terminals full bore on my $20/mo Grok sub and I'm at 9% for the week. Codex or Claude would easily have hit 5-hour or weekly limits.
I'm really not burning tokens fast enough. I use Claude a lot, daily, and have yet to hit my a ceiling with my Max/100 subscription.
Maybe because I like to verify its outputs and spend a lot of time iterating to get better outcomes. Presumably if I just let it "do its thing" I'd burn more tokens and "get more done" but I'd lose my grasp on what's in the code base.
I can respect if you say you hate their guts. Everyone has their worldview. But over moral or ethical stand? You don't have any if you're using Chinese models, or fly Middle East airlines, or countless of other products. Don't delude yourself.
once my codex/claude weekly limit was gone, i gave it a try. It was surprisingly good, not dumb in any way, and fast. I now require it as a part of 3-of-3 quorum with any codebase change.
I'm only being forced to use it at $WORK since some people overran their Cursor bill, so everyone gets Cursor Auto enabled by default which routes to Grok 4.5.
I've tried it on my "let's run every model in parallel and see which finds more edge cases" type of tasks, and Grok 4.5 was really behind Opus/ChatGPT but ahead of Gemini - despite having a strong showing on benchmarks.
That makes me really skeptical of it being GPT5.6-tier, much less Fable-tier, based on some of these benchmarks alone. But I'll test here shortly.
> I have never met a single human being who uses Grok for coding
Me too. The only people I ever saw using grok were using it by accident as they used copilot in auto mode and noticed some prompts were thrown it's way.
I recently decided to get an AI subscription and evaluated Grok vs chatgpt. Went with Grok because it's all-around good enough at day to day stuff, integrates with my Tesla, and the image/video generation is great. Kids love whimsical videos of them riding dinosaurs.
Feedback: I'd like Grok to have more connectors (I see OpenAI just added Apple Health, that would be nice to have, and I wish it could read my Onenote notebooks) and for existing ones to be improved. I gave it access to my gmail and asked it "what was my last electricity bill?". It failed to find it, even when I told it the exact subject line to search for. Something about not getting any data back when trying to get the email contents.
Grok doesn’t have those features, and people who like to make adult content have been complaining for a while how Grok has made it a lot harder to do so.
Interesting. Grok 4.5 is a capable model, although not quite at Fable/Sol levels. Will be interesting to see how this holds up. Musk appears to have made a savvy choice buying Cursor's data.
I often wonder if there's a chance, even if minimal... that they stole the weights of the Anthropic models they run on their datacenter... or are actively destillating it.
I think the more likely explanation is that the Cursor data they effectively acquired for $10B was extremely valuable for their training when combined with the insane number of GB300s xAI has for training.
I’m perpetually surprised people feel comfortable sending their code to an LLM that was made to call itself MechaHitler. I wouldn’t touch grok or now cursor with a 10 foot pole
Agreed but there were similar controversies with how OpenAI was generating images. The only pass is it was the early days of chatbots and this stuff is so non-deterministic and experimental.
This was the change Grok's team made, that was later reverted:
> - The response should not shy away from making claims which are politically incorrect, as long as they are well substantiated.
Cursor, since Grok 4.5, has had an incredible deal for frontier level models, their subscription now goes way further than OpenAI or Anthropic. Even on their lower tier plans you can use a lot tokens on their of their first party models (Grok and Composer) and not really run out comparatively. Combine them with an orchestrator and implementor type setup and it goes even further.
I believe they are the only western provider that has Kimi K3 on a subscription plan today as well. I would love to ditch Anthropic and be on Kimi if there were a subsidized plan like that with ZDR
GitHub Copilot does have Kimi K3.
What’s the multiplier? GH copilot nerfed their product so badly that I unsubscribed.
Opencode have it in their subscription
kimi k3 credits end in just a few sessions. Only Grok models allow generous use in Cursor Pro/+
GabAI has KimiK3
Can you explain what you mean? These days courtesy of an addictive reset game OpenAI is playing, I can't find anything with frontier intelligence that's more cost efficient...
How does Grok 4.5 compare to Opus >= 4.8 though?
I'm willing to pay 2x for a 10% smarter model. Intelligence matters that much (because 10% smarter probably saves, on average, several hours of human time).
It's a bit worse.
I haven't tried so it's pure speculation based on benchmarks, but I'd assume Grok 4.6 is around Opus 4.8 in real world use, but clearly below Opus 5.
I've found Fable 5 to be so much better than 4.8.
For building a full stack custom CRM and media pipeline tool with video conversion, transcription, and indexing. Supabase, AWS, Meili, NextJS, GCS - lots of surfaces and planes.
4.8 basically couldn't do it, I abandoned the project as the fallback was, "current business processes".
With F5 it's been 4 weeks and almost ready for production release.
Goes even further to exfiltrate your data, yeah.
That would be Muse Spark Contributor Tier. 12-21x price reduction at the expense of your digital existence.
I'd be the first model I'd reach for if I was providing a free service to AI gooners though. Serves them both right.
SpaceXAI is the only frontier model company that had its own compute/date centres and soon chip making factory, I think they will pull ahead with cheaper tokens similar intelligence and better harness/tools. Grok build is 2-5x faster than Claude Code in my opinion.
Thanks for your opinion, Elon.
file this under never gonna happen (this folder is piling up...) :)
I have never met a single human being who uses Grok for coding
I have. He was using it due to philosophical reasons the same way many people have philosophical reasons for avoiding it. I don't know how many people are like that, but it's not exactly where you want to position your product if you're a business.
Personally - and I know I'm not alone with this sentiment based on comments I see on this site - I wouldn't touch Grok no matter how good or cheap it is. I don't trust Elon and I don't want to give another dollar to the world's richest person who turns around and uses the money to interfere with elections. The guy I know uses it for essentially the same reason I won't use it.
> the world's richest person who turns around and uses the money to interfere with elections.
I'm no Elon fanboy but to be fair a lot of the richest people in the world heavily fund and endorse political candidates, parties, PACs, lobbyists, etc across the political spectrum. Elon was just far more public about it and it was covered much more.
This statement doesn't go far enough given Elon's direct and hands-on involvement with DOGE and the 2024 elections. Very few of the richest people of the world are personally entangled in meddling with government agencies directly, for example.
> Elon was just far more public about it and it was covered much more.
Most importantly, he simply backed the “wrong” side for most of the people who bring this up.
> MADISON, Wis. (AP) — Billionaire Elon Musk likely broke Wisconsin law when he promised to hand out $1 million checks to voters in the 2025 state Supreme Court election, a bipartisan panel has found.
> The Wisconsin Elections Commission last week referred two complaints to the Brown County district attorney’s office, which can choose to bring criminal charges over violating the state law against election bribery. Prosecutors have 40 days to report back to the commission.
https://apnews.com/article/elon-musk-wisconsin-election-mill...
Versus Sam,Dario or the CCP? Im all for running local models but im sor far from being able to pay for a large model hardware setup. My strix halo box is like driving a beaten up vespa when the frontier models are Ferraris.
>who turns around and uses the money to interfere with elections
thats a very dumb reason considering all rich people do it, most are just not as open about it as Musk
There's a big difference between quiet donations to a PAC - not that that's good either - and what Elon did. He literally paid for votes, likely in violation of the law. He poured more money into U.S. elections than anyone has before. He and Trump both made strange, cryptic statements about Elon's role in Pennsylvania with the voting machines that has caused people to reasonably wonder if they somehow manipulated the election. Whether he did or not, the innuendo alone is not ok. Then he did what he did in Germany. Don't even get me started on DOGE or his Starlink shenanigans in Ukraine.
That's before we even start talking about the models themselves. He claims to want "unbiased" models, but he very clearly has a distorted view of the world and has repeatedly demonstrated a desire and willingness to bend the world to his will. I don't want to use a model that is so obviously suspect. Not to mention, his models repeatedly produce racist, Nazi-like propaganda.
IMO, he is, at best, a clueless amateur masquerading as an expert and running into problems a more careful person manages to mostly avoid. At worst... well, you get the picture.
"Everyone's doing it" presented with no evidence is just you trying to make yourself feel better for: - using from - driving a Tesla - voting for Trump
It's a really, really easy line to draw in the sand: don't support openly corrupt individuals.
It's absolutely true there's money in politics. To call all money in politics equally corrupt because Bernie got a dollar to have dinner with someone vs Elon effectively directly buying votes....
A complete lack of nuance here. And I wouldn't be surprised if corporations / super rich WANT you to think like that. The more defeatist the mentality becomes the more we just accept whatever they do next.
A bunch of SWEs at my work use it as their primary model.
We have Claude, ChatGPT, and Cursor with essentially no cap on spend (top guy is spending over 10K a month on AI at API prices), and he hasn't had his hand slapped.
So it's not like they are using it purely because it's cheaper.
I think people like to use it for its speaking style, pretty solid performance, and its speed.
That's pretty surprising. Idk about Grok 4.6, but Grok 4.5 was clearly below Fable, Opus 5 and GPT 5.6 Sol.
It's much faster, so if you're not doing something cutting-edge, or you're doing the planning yourself and just using the LLM for implementation, the speed benefit outweighs the extra smarts of Fable/Opus5/Sol.
I had a security incident the other day and Grok was the only model that would help. Claude and GPT refused on ethical grounds and only gave general advice. In an emergency, I'd only trust Grok. However, that's the only time I used Grok for coding (since Opus 4.8 it would take a lot to get me to switch away from Anthropic)
With Claude I start having to limit the context I give it about my problem in case it trips up the safeguards. :(
I use. I used to be a Claude user. Since trying Grok 4.5 and especially Grok 4.6, I don't want to go back to Claude any more (I have early access to 4.6).
Grok is 3x+ faster than Claude and I can't tell the diff in engineering work quality. As an engineer, speed is important to me.
For $30/month, I'd expect it to have higher usage limits than Claude Code and Codex.
It really does. I felt like I could have spent $1000+ api token on claude for the amount of work on my $30 grok subscription.
An hour in, I've been running four terminals full bore on my $20/mo Grok sub and I'm at 9% for the week. Codex or Claude would easily have hit 5-hour or weekly limits.
I'm really not burning tokens fast enough. I use Claude a lot, daily, and have yet to hit my a ceiling with my Max/100 subscription.
Maybe because I like to verify its outputs and spend a lot of time iterating to get better outcomes. Presumably if I just let it "do its thing" I'd burn more tokens and "get more done" but I'd lose my grasp on what's in the code base.
I refuse to use that product because of the parent company.
100%, it's an easy pass given that it is always playing catch up.
I can respect if you say you hate their guts. Everyone has their worldview. But over moral or ethical stand? You don't have any if you're using Chinese models, or fly Middle East airlines, or countless of other products. Don't delude yourself.
Chinese models are open, Grok is not.
once my codex/claude weekly limit was gone, i gave it a try. It was surprisingly good, not dumb in any way, and fast. I now require it as a part of 3-of-3 quorum with any codebase change.
Does it matter? Why turn it into a popularity contest?
Folks working in US govt tend to, based on convos I've had with one such person.
They make good stuff
Not the best endorsement given the current US gov
it only very recently became competetive, if they proceed with improvements (and beating others on price) their share will grow
I'm only being forced to use it at $WORK since some people overran their Cursor bill, so everyone gets Cursor Auto enabled by default which routes to Grok 4.5.
I am using it for code reviews, and it regularly surfaces stuff that neither Sol or Fable do.
the new models are quite good, give it a shot
Hi! Grok’s worked quite well for my use cases…
It’s also a great deal!
I use it because it's cheap and good enough
I've tried it on my "let's run every model in parallel and see which finds more edge cases" type of tasks, and Grok 4.5 was really behind Opus/ChatGPT but ahead of Gemini - despite having a strong showing on benchmarks.
That makes me really skeptical of it being GPT5.6-tier, much less Fable-tier, based on some of these benchmarks alone. But I'll test here shortly.
It's still not as good as GPT5.6 or Opus5 but it's better than KimiK3. Good job xAI team.
> I have never met a single human being who uses Grok for coding
Me too. The only people I ever saw using grok were using it by accident as they used copilot in auto mode and noticed some prompts were thrown it's way.
I saw far more people using Mistral than grok.
Hello. Nice to meet you.
Grok is not the best model around, but it's decent. It gets the basic job done at a low price. I don't think it can advance frontier Math, yet.
Probably can't advance frontier math yet, yeah. But please let us know other places you want to see Grok improve for future models!
I recently decided to get an AI subscription and evaluated Grok vs chatgpt. Went with Grok because it's all-around good enough at day to day stuff, integrates with my Tesla, and the image/video generation is great. Kids love whimsical videos of them riding dinosaurs.
Feedback: I'd like Grok to have more connectors (I see OpenAI just added Apple Health, that would be nice to have, and I wish it could read my Onenote notebooks) and for existing ones to be improved. I gave it access to my gmail and asked it "what was my last electricity bill?". It failed to find it, even when I told it the exact subject line to search for. Something about not getting any data back when trying to get the email contents.
More deepfake porn creation please! Oh yeah, and for children too!
Grok doesn’t have those features, and people who like to make adult content have been complaining for a while how Grok has made it a lot harder to do so.
That's why I'm asking for them!!!
https://news.ycombinator.com/item?id=49274027 what happened here?
no dupe?
Seems the cache read pricing almost doubled from $0.30 in Grok 4.5 to $0.50 in Grok 4.6.
In my experience in heavy coding sessions most pricing is just cache read and cache write like 80% of my token bill.
Why is Grok so much cheaper than Claude or GPT?
Elon owns a lot of compute
Interesting. Grok 4.5 is a capable model, although not quite at Fable/Sol levels. Will be interesting to see how this holds up. Musk appears to have made a savvy choice buying Cursor's data.
I often wonder if there's a chance, even if minimal... that they stole the weights of the Anthropic models they run on their datacenter... or are actively destillating it.
I think the more likely explanation is that the Cursor data they effectively acquired for $10B was extremely valuable for their training when combined with the insane number of GB300s xAI has for training.
Cursor was 60B. The 10B number was the breakup fee if the deal fell through.
$60B in SPCX stock, which is sort of magic money.
> ... or are actively destillating it.
I just assumed every model manufacturer is distilling from the frontier models. If they aren't they are definitely trying to do it.
I’m perpetually surprised people feel comfortable sending their code to an LLM that was made to call itself MechaHitler. I wouldn’t touch grok or now cursor with a 10 foot pole
MechaHitler was something that existed only on X's grok chatbot, due to a one-line system prompt change they reverted after half a day.
That's different than using Grok as a model for coding.
"I only did the Sieg Heil once"
It's true, but I do worry about governance when it comes to these models. That shows a surprising lack of discipline in their deployment pipeline.
Agreed but there were similar controversies with how OpenAI was generating images. The only pass is it was the early days of chatbots and this stuff is so non-deterministic and experimental.
This was the change Grok's team made, that was later reverted:
> - The response should not shy away from making claims which are politically incorrect, as long as they are well substantiated.
https://github.com/xai-org/grok-prompts/commit/c5de4a14feb50...
oh, it was only due to a "one-line system prompt change"? well okay then! Here I was thinking it was multiple lines! How foolish was I??