Grok 3 VS Gemini 2.0 vs Perplexity VS Qwen 2.5 Max: Who Wins?

grock 3 versus Gemini 2. 0 versus perplexity versus quen 2. 5 Max who wins let's find out today we're going to get

straight into the testing and the first thing we're going to test is the image generation on each so I'm going to say inside Gro 3 picture of a UFC style Gladiator battle between AI robots in a packed Arena we'll do the same on blex now by the way to generate images inside plexity we can just go to generate images in the bottom right and then we can select photograph and that will start generating like so inside AI Studio this is where we can access Gemini 2. 0 so let's go with Gemini 2. 0 Flash the newest model from Google and we'll plug in the same prompt and then we'll run this inside quen 2. 5 Max by the way these are all good models that I like they're all very underrated like for example not many people talk about using quen 2. 5 Max as well totally underrated tool that's actually really powerful like you can see so let's see what we got here the images from grock 3 this one is pretty cool to be fair like that's interesting like a scene from a movie almost looks very cool let's have a look at plexity pity is not known for image generation honestly I wouldn't say that's the best image I would say that's a pretty cool image look at some of these images here to be fair all this beat chat GPT when it comes to image generation Geminis is pretty cool and you got quen 2. five Maxes as well which is really cool so if I had to rate them I'm going to go with perplexity first actually I really like this style of you got the Coliseum in Rome falling apart fire in the background these two images really this one is pretty cool too it just lacks a little bit of character that's what I would say but Gemini's image generation there is good I like GRS but the characters are just looking away from each other which kind of feels too weird and then finally we have the image from quen 2. 5 Max it just feels a little bit weird like they've got multiple arms the character the people in the background look a little bit cartoonish I'm going to put that last next up we're going to be using the next up so Gro came in third right there let's test the Deep

search method of research so we're going to say tell me about the latest Brock free release tur into a beautifully designed Report with Emojis headlines and subheadings we'll do the same inside pxi now one thing you have to note here is when I'm doing this is we'll be testing the research me methods from all these out however what I would say here is like Gro and perplexia are actually built for deep search methods right whereas inside Gemini the way that we're going to do this is with the Deep research so they have a deep research AI research system so we have to select 1. 5 Pro go to Gemini then we actually have to upgrade to Gemini Advanced and we're going to take one for the team here and sign up so if we go inside Gemini now we're going to go to 1. 5 Pro with deep research and then we can test this out B bear in mind this is only available on the paid plan of Gemini so perplexity does it for free which is a massive win grock 3 does it for free as well massive win there Gemini you actually have to pay for deep research reports but we're going to go with that anyway we'll start the research I do like the way it's planned out and then you approve it before you get the research report back now Quint 2. 5 Max is not really built for this but what we're going to do is we're going to say okay tell me about the latest Gro 3 release using the web search feature inside quen and we'll see what we get back there now we've got the results back from grop 3 so that was pretty quick here it's talked about all the headlines Etc and it's planned out and shown us each part of the process right so first we're thinking exploring the user request researching G 3 Etc then if we go down to the key points got some nice bullet points there super useful we've got the release overview it's really easy to understand this it's got accurate dates the data is useful it's got some nice sources nice benchmarks table look at that that's beautiful so we got the benchmarks again three grow fre Mini gp40 Gemini 2. 0 deep seek and Claude Sonic and it's really easy for us to compare inside this research report I'm going to say that's pretty good stuff in the only thing I would say about this is there's no real personality inside there it's not that interesting to read if we have a look inside perplexity this has a lot more sources so you can see here we've got like just in the first paragraph we got six sources nice bullet points good Benchmark comparison of Gro 2 versus Gro 3 and then it's got the key benchmarks again in a nice little table comparing GPT 4 2. 0 pro and DC G one then it's also talked about some real world applications the long-term vision and future development road map right so what I like about this versus say Gro is that if you look at grock's report it's not that interesting to read it's not that well formatted so I'm actually going to go with perplexity winning on the Deep research method then Gro 3 so far still waiting for Gemini to come back to us to be fair quen 2. 5 Maxes report is really good like the way it's nicely formatted it's easy to read you've got the sources on the left on the right hand side here there's no tables or comparison of benchmarks but it does have the latest date and timelines here so it knows what's happened recently which is good what I also think is great about Gemini though to be fair is it can actually look at YouTube videos you see how it Source YouTube 1 2 3 four five times inside the content I think that's great because for me personally like I don't really read blogs anymore but I will watch YouTube tutorials on new releases just to get the latest updates from it to me that's a really positive sign but let's see what we get back the other cool thing here is like inside Gemini you can also open this up in a Google doc and then you've got this seven page report like you can see right here on the latest updates from Gro I don't think enough people have talked about the Deep research release from gini but it's actually looking really good like you can see nice and easy to read lots of good data there I do not like this table though look at this so it has the benchmarks Gro 3 competitor 1 competitor 2 but there's no scores like deep seek the actual report Port here is not useful at all so it's failed on the table of research right there so for example like most of the table is blank you're not really sure what's the score the Deep SE called GPT 4 got and also this is not the most relevant model to Gro 3 when we're comparing science right we want to be looking at Deep seek R1 versus gr 3 not deep seek version 3 applications to use cases is decent lot of good sources there overall not that overall though I wouldn't say that's anywhere near as good as perplexia right so perplexity wins a research report grock 3 comes in second Gemini comes in third and then quen 2. 5 Max comes in last and Gemini would have done a lot better but the table comparing the key benchmarks across all the AI models isn't useful at all right if you had to turn that research report in it's just

going to be a fail so now we're going to test grock 3 versus arrest to write a python script of a ball bouncing inside of spinning Tesseract we'll plug that into Gro same into complexity Gemini Advanced and then finally into quen 2. 5 Max as well and let's see how we do here so one of the things that I do like about the UI and Gro is it actually gives you a countdown timer so it tells you how long it's been thinking for plus all the updates along the way it actually looks like pxi is totally failed so it says here's a python script that creates an animation of a ball bouncing there's no script there oh python code mate what are we going to do with that so it looks like plex's already failed let's have a look at Gemini is giving giv us python script as well using 2. 0 Flash and then quen 2. 5 Max as well is building it out also I like quen 2. 5 Max cuz it has an artifacts feature but it seems to be really buggy look at that is not what we're looking for right there so let's see what we get back from Gro we're still waiting for it to load it's been 81 seconds but we can take the code from Gemini that seems to be really fast and just test this out if you ever want to preview python code online trinket. io is pretty good I'm just going to plug in the python code and then we'll run this is a code from Gemini let's see what we got so it's sing up now it's not what we're looking for right we got a DOT bouncing around inside the Tesseract but we can't actually see the Tess act not particularly useful now we got the code back from Gro so let's test this out and Gro actually did it right look at that ball bouncing inside a spinning Tesseract Gro is the only one that won there right so Gro actually gave us a working model gerini it gave us a bouncing ball but it didn't give us a tesseract so 3D rotation of the 3D shape same with quen so failed cuz look at that there's no real you're not really that's not what we're looking for at all and then perplexity just didn't give us anything but it stated it did so the only one that really did it probably was gr 3 there using the think mode next up now

we're going to create a onepage website for our new service video ranking service so I'm going to plug this in it's got some information about me who I am Etc we Ed the thinky mode as well just cuz we're coding again let's ask Gemini to do the same we can test this out with quen 2. 5 Max artifacts feature and then we'll use the reasoning mode with deep with 03 mini I think is best for this I used to use deep SE G one a lot but it just doesn't seem to perform when I've done recent tests versus before and that's the thing about this whole game right there could be an update this week say chat GPT or open AI releases GPT 4. 5 and all of a sudden that's the best model in the game that's how fast this moves it's literally every 7 days this is why I'm constantly posting updates there's a new tool that's better than the previous one so quen 2. 5 Max is come back with the goods now let's pull this full screen we can have a look here so I love the fact that you can just preview the web page directly but as you can the design is a little bit basic but it is all personalized to us it's got the popup as well it's okay it's not bad it's okay I wouldn't host it like that I probably have to go back and forth with quen 2. 5 Max a lot more now I've got the code back from Gemini let's test it out we got the JS as well that is looking a lot better right so we got the popup we got the landing page 10 times better than what you get back from quen 2. 5 Max the design is nice the font is nice we can't X off the popup but if you look at the actual design in the background it looks 10 times better all the Emojis and everything like that right exact same prompt but you can see a huge difference between what you get back from Gemini Advanced versus what you get back from quen 2. 5 Max now I've got the code back from Plexi so let's test this out now this is using gpt3 mini so we've grabbed the CSS and then we grabb the HTML as well and I really don't like the design of this if we look at this again if we compare it versus a Gemini so Gemini's code is right here looks great like this would be ready to publish honestly whereas if we look at the code from perplexity using gb03 mini you just wouldn't publish that right look how the buttons are overlapping the sentences the content super Bland white on white text on white background when has that ever been a good idea I would say that's quen 2. 5 Max and perplexity coming in last right there let's check out Gro 3's content so we've got the code right here I also like the fact that it's actually just giving us a complete code inside HTML so we've got the CSS and the JS built in it's not bad I would say that's better than quen 2. 5 Maxes but worse than Geminis right Gemini is still dominating so for code in bu a long way Gemini Advance 2. 0 flash thinking experiment m al is winning the race here if you want to get free access to this by the way just go to AI studio. google. com and then from the list you can select Gemini 2. 0 experimental Pro right there or Flash thinking experimental and you can get access to these models for free you don't have to pay for Gemini Advanced but yeah it blew everything out the water and quen came in third from what I can see is very basic and with perplexity coming out last now inside gro we going to run this prompt for creating content like you can see so

we're going to create an SEO optimized article for this keyword equals gr 3 AO let's plug that in see what we get back and we'll run the same thing through Gema I'm just going to use 2. 0 flash for this CU you don't need a thinking model to create content we'll do the same inside quen 2. 5 Max I'm actually going to allow web search as well so it can create the content and then we'll do the same inside perplexity we'll Select Auto for the search and then we'll give it the same prpt all right so Gro has already come back with the goods so Gemini is really fast so actually planning out the content then the outline the only thing I don't like about the content from Gemini here is number one it's not really SEO optimized right so if you look at the content section it says let's get straight to it you're probably wondering is Gro 3 ASO going to make my job obsolete or maybe how do I even use this stuff he not very useful right if we look at the content from Gro it's got a nice heading nice introduction it's all personalized to me he quite humanized let's have a look at quen 2. 5 Max this is more like it right this is what you're looking for so you got a nice little header it's hook and then it gets straight into the content now the other problem with Gemini's content here is it's very short there's no headings it's not nicely formatted and it's super basic perplex content is pretty good as well says Gro fre AI the ultimate SEO Game Changer here's the content this is super subjective right so what I might like in an article you might not like but what I would say here is like quen 2. 5 Max he's winning pexd comes in second grop 3 comes in third I just don't it's just a really long intro with loads of fluff that doesn't help so I don't and then Gemini is just full of like random bullet points and there's not even a there's not even an article header right there so Gemini is coming in last for that little bonus tip here right if you're going to create content I would actually recommend clae out of all of these I think it wipes the floor of all four models so if I plug in the exact same prompt the content is way more humanized it reads better it's nicely formatted it's created on the canvas it's free to use Claude And it it's by far the best content writer that I've seen out there so let's compare who won

who didn't and round up so we had four contenders Gro 3 Gemini 2. 0 perplexity and quen 2. 5 max round one for image generation perplexity 1 and C in first quen 2. 5 Max came in last just a little bit too cartoonish and we proportions whereas flexi did the Epic Coliseum battle and great atmosphere for deep research perp Plexi also won that had the most sers the best structure and great benchmarks GR 3 actually came in second so it's still a decent option and this and when it came to python coding Gro 3 was the only one to generate a working Tesseract model perplexity came in fourth so it's not great for python coding grock 3 is probably the best option for that along with Gemini 2. 0 was closest behind but honestly none of them compared to Gro 3 website creation Gemini 2. 0 by long way was the best clean design well structured best overall the rest were nowhere near as polished as Gemini 2. 0 and then for SEO content writing quen 2. 5 Max was actually the best but plexity came in second it was still strong Gro 3 was a bit fluffy and then Gemini 2. 0 just had random bullet points it was short and it had no headings who won what is a tie between SE and Gro 3 so grock 3 and pxi draw at the top perplex is great for image generation deep research grock 3 is really good for python coding bear in mind Plexi and Gro 3 free models whereas for example for coding with Gemini 2. 0 you need to upgrade to Advanced long with deep research so there we go let me know what you think in the comments so

thanks so much for watching if you want to get all the proms from today make sure you join the AI profit boardroom Link in the comments description this comes with an awesome classroom and loads of cool stuff including email and content automation social media AI agents web Automation and scraping Q& A call recordings a crash course on how to save hundreds of hours and make money with AI and also all of my best SS I've created right every single day we update this with one to two newps and if you go inside the gro 3 versus chat GPT you can you'll find all the prompts from today additionally this comes with like loads of SPS on like how to make $100 for deep seek na10 agent workflows a 3our 45 minute course on exactly how to use deep seek like everything you can ever imagine is all inside here and it's being updated daily along with a community of 189 business owners all interested in growing their business and scaling saving hundreds of hours with AI by the way we did an update 9 19 hours ago and it was the third update in the last 3 days right so we just constantly posting new stuff inside here that you can learn from and additionally if you go to the calendar section you can actually jump on weekly Q& A calls with me next one is in less than 48 hours so make sure you sign up now so you don't miss out on that link in the comments description and if you want to get a free one to1 SEO strategy session that shows you how we websites from 0 to 145,000 Vis this month and generate hundreds of thousands of dollars in sales on auto wallot feel free to get that on this free link building acceleration session you get free SEO domination plan so you get a Custom Tailor link building plan to generate more lead sales and profits from your website you'll discover the secret SEO link building answer any questions you have one to one you learn the best link building strategy for your website pleas add to quickly ranking competitors link building and how to standco traffic based on what's working for us and our happy clients like you can see right here feel free to get that link in the comments description appreciate you watching byebye