Skip to content
Elons Xai Grok Gets 2X Faster After Devs Re-Write Code in 3 Days episode artwork

Nov 5, 2024 9 min

Elons Xai Grok Gets 2X Faster After Devs Re-Write Code in 3 Days

from Hard Fork AI

XAI's AI model Grok 2 received a significant speed boost after developers rewrote the inference code stack using SG Lang in just three days. This open-source system executed complex language model programs and achieved higher throughput. The developers, Liam Zhang and Sied Melik, accomplished this feat, resulting in Grok 2 being two times faster and slightly more accurate. The success of this update sets a precedent for improving AI models without the need for retraining or expensive infrastructure. Grok 2 has also secured the number two spot on the Limsis chatbot arena leaderboard, with impressive user ratings. Get on the AI Box Waitlist: ⁠⁠⁠https://AIBox.ai/⁠⁠⁠ Join my AI Hustle Community: ⁠https://www.skool.com/aihustle/about⁠ 00:00 Introduction to XAI and Grok 02:14 Code Rewrite and Speed Boost with SG Lang 03:13 SG Lang: Efficient Language Model Execution 05:34 Grok 2 Ranks Second on LimSis Leaderboard 07:26 XAI's Progress with Grok and Image Generation

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

Elons Xai Grok Gets 2X Faster After Devs Re-Write Code in 3 Days

0:00 9:17
of matches

Transcript Auto-generated, may contain errors

The team over at XAI continues to blast out new AI updates at a blistering pace I'm actually very impressed this was a company that I mean I would say like never underestimate Elon Musk and other people love him or hate him Whatever don't underestimate him. He's got a lot of big companies that have done a lot of impressive things But at the same time I was I just felt like XAI was coming a little bit late to the party I felt like Elon Musk has kind of missed his boat with open AI and kind of stepping away from that right before it went parabolic And so starting XAI was wasn't a hundred century ago I'm I've been quite impressed with the progress that Grock has made and this week something very interesting happened And that is that Grock too, which is their big model It got a huge speed bump which is interesting because developers essentially rewrote all the code in three days And then got some massive speed bumps. I'm gonna talk about how this happened What is looking like what you know how they're able to do this before I get into that I wanted to mention that my podcast course you might have heard I launched this but we could go right now I'm trying to get more reviews and testimonials for it. So for this week only I am doing for AI chat listeners I'm doing a discount code normally things $300 if you're interested in starting a podcast growing it all the strategies I use to scale my podcast over four million downloads We'll be available in that course so go get it if you're interested in it do it this week I'll obviously increase that in the future But go get it the link is in the description and the promo code is AI chat My only thing I ask is if you are gonna take advantage of that to please shoot me over a review or testimonial on LinkedIn If you love the course, so let's talk about Grock the thing that I think is really interesting here is that in order to get Grock It's of course the LLM It's on X so you have to have at least $8 a month subscription Which to be honest not bad because I should use $20 so if you just get an X premium subscription or you do get Grock And it's essentially the same thing You know check it be slightly better.

I won't discount that but Grock is just added image generation impressive and now this new speed boost So they have kind of like type of TIPT with GPT mini or whatever they have Grock 2 and then Grock 2 mini Which is a little bit less powerful, but it's faster so you can use both of those But this has all been due to something that has happened over the last three days were some developers at X AI were able to rewrite the inference Code stack completely in three days. So a lot of this comes from a tweet over on X by Igor Babushkin Who posted and said quote Grock too many is now two times faster than it was yesterday the last three days In zigong and maliki said re-wrote inference that from scratch using sg laying This has also allowed us to serve the big rock to model which requires multi-host inference at a reasonable speed both models Didn't didn't just get faster, but also slightly more accurate stay tuned for further speed improvements What I think is really interesting here. This is something that a lot of these AI models can do It's not like they had to retrain the whole model or spend hundreds of millions of dollars to do do that They essentially just rewrote the inference stack using sg laying something that you know virtually yeah a lot of people could do It's a sg laying is It's on github so very very interesting So the two developers that did that are Liam Zhang and seed millike and they were able to use sg Language is actually an open source Apache to licensed It's you know a highly efficient system for except for essentially what it does is it executes complex language model programs But it can get up to 6.4 times higher throughput than existing systems and so for Grock 2 that was able to get it two times faster Which is quite impressive sg laying was actually developed over at Stanford University a bunch of researchers and the University of California Berkeley Texas A&M University and then Shanghai's Geotong University. So all these big universities working together This is something that I love essentially were able to use technology to make these AI models more efficient and faster without having to you know Do like a whole retrain and spend millions of dollars and you know data centers and GPUs and like all the crazy infrastructure that's required They're just rewriting code And you know it's kind of an undertaking but like two developers for Grock did this in three days Which is I mean obviously these are superhero developers, but two developers three days and this thing's two times as fast that is Absolutely phenomenal and that's an amazing precedent.

So great work over to the team there and I hope to see this world out other places So the same system that was currently supporting a bunch of different models There's llama mr. All which are all compatible with open weight and API based models So this is gonna be very interesting to see who adopts this next and how's it goes So the last thing I want to bring up is you know, what are the results? How is Grock 2 and Grock too many doing does I don't even use this so there is a third party? thing called limsis chatbot arena leaderboard and it's essentially ranking chatbots people vote on them and rate their performance it's independent third party and Impressively after all of these upgrades Grock 2 has now secured the number two spot after Chachuk T40 And they have a really impressive score of 1293 this is based off of over 6,000 votes of people testing and reviewing this and so yeah very very impressive made some big updates And it seems like this is getting better the last thing that I wanted to mention is that It didn't just get faster.

It also got better. They said that the response is slightly better So I thought that was also very interesting It'll be interesting to see how how this changes and other people adapt this in the future So definitely come to keep your eye on and I'll keep you up to date on everything happening with Grock If you haven't tried it already, I would highly recommend making an account on X or you know getting it if you don't have one and Giving Grock a try it's quite an impressive model. I've been quite impressed with it and especially with their new image generator I think their new image generator is better than I want you to get out of dolly on chat GPT So I'm exclusively having my having my production team use it for thumbnails and things like that instead of dolly from Chachuk T Which is a little too cartoony the AI generator on Grock, which is a an API they have on there But it is it's much better images seem photo realistic. So that's fantastic if you enjoyed the episode today I would love it if you could leave a review for the podcast I hope you all have an amazing rest of your day And I also wanted to mention that if you are interested in ways to make money with AI or start a side hustle with AI You can join the AI hustle school community This is school community I launched with the co-host of AI my hustle podcast Jamie We record exclusive content every single week talking about different ways that we are personally making money from AI tools last week We released an episode or I guess this week we posted an episode about the Amazon influencer program and a new initiative that they're doing last year I made $12,500 from this initiative in about three days of making videos They had an incentive program and they're running incentive program again this year So if you're interested in making money from AI and AI tools go and join the community It's $25 a month we'll increase it to $50 a hundred a month of future But for now $25 a month we'd love to have you as part of the community You can get a link in the description to the AI hustle school community Hope you all have an amazing rest of your day

Frequently asked questions

How long is this episode of Hard Fork AI?

This episode is 9 minutes long.

When was this Hard Fork AI episode published?

This episode was published on November 5, 2024.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this Hard Fork AI episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.