How To Use New Google Gemini (Gemini AI Tutorial) Complete Guide With Tips and Tricks
LARRUZ walks through Google Gemini's core features: choose Standard or Advanced (Advanced handles coding and complex work), enable extensions to search Gmail and YouTube, upload PDFs and images with 1M token support, convert email results into tables for Google Sheets, run code in-browser, and use the G button to fact-check claims. Image generation is blocked in EU/UK but works via USA VPN.
Chapters
So, since Google Gemini actually got an update, I wanted to make an updated tutorial on all of the things that you can now do with Gemini, including some tips and tricks, because there are a lot of things that people actually don't know, and Google has put some links in places they shouldn't be. So I'm here to just give you guys the full rundown on how you can actually use this and be really effective. So one of the first things you should do when coming into the app is, of course, you want to go on to your settings and then you want to just hit dark mode. Now, I did this because in previous tutorials, I know that a lot of people prefer watching them in the dark mode because it's just way too bright. And you can do that down at the bottom right there. Now, in addition to that, what you also want to make sure is, and this is something that I've made mistakes from before, is that when you're starting to use Gemini, you always want to make sure that depending on your task, you either use Gemini or Gemini Advanced. Now, this does seem quite basic, but sometimes when your membership does get canceled, Gemini doesn't actually explicitly tell you this, and. And it will just automatically switch back to Gemini. So if you see just Gemini right here, always just double check which one you're on. Gemini Advanced is for more ambitious products that require more cognitive ability. So, for example, coding and just things about like physics or whatever, stuff that's a little bit more advanced. But Gemini Standard is stuff that's for very easy things like translations, messages, and emails. So that is the key difference there. Now that you have this, I'm going to show you guys some of the cool things that you can do. So one of the things that you also want to do quickly before you start is you want to click your extensions and then down here, what you want to do is you want to make sure that you've got these all ticked. Now, I've got these all ticked because I want to be able to use these and show you guys what you can do with this later on in the video. So tick these because this is going to give you more access to Gemini. So one of the first things you can actually do is you can type in at right here. So when you type in at right here, you can see that we have several different apps that we can connect with that we can use. So, for example, if I wanted to connect with my Gmail, I can then use this to search for a relevant email. So, for example, what I can do is I can say, hey, Gmail. So I just want to click this button here, just click Gmail. And then I'm just going to ask it what properties have been emailed to me. Deals are on the table. What deals are on the table. Okay. So I'm just literally going to ask what properties have been emailed to me and what deals are on the table. So now it's going to access my emails, search through this, and I think that this is really, really effective because when you know what Gemini is able to do, you can combine that with certain things, and then you're able to get a comprehensive overview of your Gmail. So it might take like, around 10 to 20 seconds. But you can see right here, it's showing me now all of the deals that I've been emailed over the past couple of weeks or maybe just last weeks. And you can see that right there. This looks really cool. Now, another thing that you can do, and this is what I would do if I was using information from my Gmail. And this is something that is not unique to Gemini, but it's a unique feature that Gemini always has had. But most people don't realize you can do this. Okay. And I'm going to show you guys how to do this again. Essentially, what you want to do here is you want to say, make you make this data into a table. Okay? So now I'm going to say to Gemini, because what Gemini can do is it can actually turn the information that you have into a table. So all you want to do is to ask Gemini, can you make this data into a table? And then you can see right here, it's managing to go ahead and put this into a table. And it says it's in a list format. So I'm just going to have to wait for it to do that. And so you can see right here that it manages to get all of this data into a quick table that I can then use. Now, I got to be honest, this is a pretty, pretty long table. You can see all of the descriptions, all of the yields, all of the prices, which is pretty cool. I really do like this. Of course, this could be a little bit better, but this is a lot more, you know, organized than this. And this is one of the ways that people are saving lots of time, because I'm sure you get emailed a lot of different things all the time. I know that literally everyone has a Gmail account. So using this and then saying, put this into a table, and I mean, you could filter this even more. You could Say, you know, give me deals that are only above 5%, deals that are only below 5%, give me deals in a certain area. And then you could do that. Now what's also cool about Gemini is that when working with tables is that you can also export this to sheets. So I can click Exports to sheets right here. And this is basically just Google's version of Excel. So I'm going to click open Sheets. And then once you're in here, you can then of course have this information and you can literally do what it with whatever you want with it. Now another thing that I really like about Gemini that you can really do, and most people don't really take advantage of this because they just don't know it exists. But you could also use the app button to use YouTube and then you can analyze certain videos. So then I'm going to paste the link in after saying at YouTube and then I'm going to say, can you tell me what this is about? And summarize in bullet points. So I'm going to literally just ask it to summarize in bullet points. And this is a really good feature if you're trying to save time. Hopefully they do add this in the actual YouTube app. But you can see right here, I've just added a video which is based on Google's IO that was really recent. And you can see right here that this is something that gives you a quick overview in bullet points. You can make it shorter, you can make it longer. It depends on what you want to do. So of course, this is something that is really, really time saving. So this is something that you're going to want to do. Now in addition, with that, I'm going to show you guys one more before I move on to some of the other things. But you can see right here as well that you can actually use Google Drive and you can actually find stuff. So if you're having any trouble finding something in your documents, you can find that document and then Google can help locate that for you with Gemini. And this is really useful because when you search with your Google Drive, you don't actually search the PDFs, you only currently search the titles of the PDF. So it's very, very useful to use this. And I'm going to show you guys how to do that. So, for example, to show you just how good Google Drive is, and this is something that is genuinely a time saver, I'm going to show you guys a PDF, I'm going to upload it and I'm going to show you guys me finding that just based on some obscure information in the PDF. So let me get you guys a PDF example. So, for example, let's say I have this PDF, okay? And this PDF is about, you know, the capabilities of Gemini models in medicine. And. And this is a pretty obscure PDF because I've saved it as a bunch of random letters, okay? So with this, what I'm going to do is I'm going to upload this to a new folder, okay? So now what I'm going to do is I'm literally just going to upload this PDF. And you can see the PDF is right here. The PDF shows that this is literally called capabilities of Gemini models in medicine. Now this is in a folder called Gemini Stuff, and you can see that the title of this folder is really strange, but that doesn't matter at all. So what I'm going to do now is I'm going to ask it something about the document. I'm going to say which document is in my Google Drive that talks about the capabilities of large language models in medicine. So you just type drive, then you just type in which file in my drive talks about the capabilities of LLMs in medicine. And I'm going to actually try and spell that properly. So now I've literally just asked it to search through my document and say, which file in my drive talks about the capabilities of LLMs in medicine. It's now going to assess the document and then it's going to find the relevant one. The reason I like this so much is because it is very, very effective at doing this. And it is something that is showing us how good it really is at retrieval. So you can see right here it says, this PDF discusses the capabilities of Gemini models in medicine. And you can see it's literally right there. Those are the items considered for the response. And I think this saves so much time because I have so many files on drive and often sometimes I misplace them. I want to know if they're deleted. And Google Gemini can literally just SC through all of your files and quickly find something that you really need to use. This is a major time saver, so I would definitely incorporate this into your workflow. Now, one of the newer features with Gemini that I really do like is what you can do is you can click this plus button here. And now we actually do have the ability to upload files. Previously we couldn't actually upload files, but now with the 1 million context length, we can actually upload files and then analyze those files. So what I'm going to do is I'm just going to upload a PDF. I've uploaded the same PDF right here and I'm going to say summarize this and give me the key points in bullet points. Okay, so this is really, really good because now we can literally just import any PDF and ask it to talk about it straight away. So right now it's analyzing, reading the documents, working the response. And this is something that is really, really good in terms of what it's able to find. And you can see it's able to give me the bullet points of this PDF, a real, real time saver. And I would argue that compared to other ones, this is really, really good because Google Gemini is focused on long context. So with PDFs, it's going to be really, really effective. Now, something that you can do that I find really, really nice with this is because you can upload so much content, what you can do is you can draw different responses and different conclusions from the data that you're getting. So for example, I'm going to upload two different PDFs and then I'm going to ask Gemini to draw conclusions between these PDFs. So I've uploaded two PDFs, one on misalignment and catastrophe and one on AI safety frameworks. And I'm literally going to say, what are the similarities between these two PDFs, so I'm literally just going to go ahead and ask it, what are the similarities between these two PDFs? And then it's going to go ahead, analyze the two PDFs, get some information and bring me a response. This is really, really good for analyzing work documents or analyzing your thoughts, organizing your thoughts. Just being able to cross reference multiple files gives you a new way to look at things. And you can say, both documents discuss the risks of generative AI emphasizing they need yada yada yada. They both highlight the potential this yada yada yada. And you can see it gives you these sources. Now of course there's many different ways that you can do this. You can prompt it to say, you know, draw something in a table about the similarities or whatever. But just having access to multiple file uploads is something that is really good. In addition, with this you can use the upload image and it's really good at image identification. For example, if I have an image, I can literally say what is this image of? And this is pretty basic, but if you want to know what car it is, you could simply ask, what is this image of? It's Going to give you a very specific one. And you can see here it doesn't just say, this features a blue sports car winding down a road. What it actually does say is it actually does say, this is the image of a Maserati MC20, a high performance sports car manufactured by Maserati. And I think that that is important because it allows you to understand actually what things are and you're allowed to ask it different things. Like knowing the make and model of the car is really important because these extra details can help you truly understand what's inside of an image. One thing that you can't do with Gemini Advanced at the moment is you can't do any images of people. I'm not sure why, I'm guessing it's just a privacy issue. But images with any faces or any people, it just simply identifies them and then it removes that at the moment. So this is something that you can't currently do, which is currently just a small limitation. So if you're wondering what things you can analyze, you can analyze locations, you can analyze plants, for example, you can analyze different species of plants. I'm not someone who's like a botanist or anything like that. This is not really my interest. But it is a plant that is one that was quite rare that I looked into. And you can see that it's easily able to analyze this. One of the key features of Gemini, I don't know why this is, but for whatever reason, it's really good at analyzing certain images and actually cross referencing that with Google. I'm not sure what they use in their training data, but as far as I know, they have really, really comprehensive training data, meaning that they're able to completely analyze what's going on in the image and really understand what's going on in there. Now something that Gemini is really, really cool for and something that I really, really like to use is the way how you can change your responses. And this is something that most people don't even know that you can do with Gemini. But. But it's something that I use all the time. So for example, I could say, can you create a story about this? So I just asked it to create a quick story. I didn't give it a much detail at all. I really didn't give it very, very basic. But one of the things that Gemini really excels at is creative writing. But that's not the tip trick here. What you can do is when you have only a creative output, which means that like your output isn't bound by things that are True or false. What you can do is you can simply highlight the text and then you can change it. So, for example, this is a story about someone who created an AI YouTube channel to make AI news. And it went really well. And we can read this story and then we can change certain pieces of text because we want the story to be different. This is extremely useful because rather than actually having to copy all of this text and then copying this going into Word or going into Google Docs and then changing it, we can do it directly here, which is going to give us a lot more control over where the story goes. And then so, for example, we can highlight literally anything and then we can modify the selected text and we can modify it with a prompt. We can make it shorter, we can make it longer, we can completely remove it, or we can just regenerate it for more creativity. So for example, I could say this is, you know, change this name to the AI grid and then I'm going to just click Enter and then it should modify that really quickly. And then you can see right here, it then changed to the AI grid like that. And this is something that's really cool. So I said right here, what I could also do is I could also make this longer. So it's now going to modify the response, it's going to make this a little bit longer, which is going to give me a decent more amount of information. So you can see it just opened up and got two paragraphs. The reason this is really, really good is because if I want to make longer pieces of text, I can make the intro shorter, I can remove it, I can change it. I can really just change everything about this. Now this doesn't work for certain things. Like for example, if you're asking information on Google, like if you're asking factual information, this doesn't work. This only works for creative stories. That's why a lot of people don't realize this, because they don't see this all the time. Now what you can also do as well is you can also change the writing style. Now this is different from this, but what you can do is you can also change the writing style. So you can make it shorter. And this is completely, overall, you can make it longer, you can make it simpler, you can make it more casual and you can make it more professional. This is something that you can do and it's really, really effective. You can do this in many different ways. But if you're trying to write something, you're trying to write an email, you want it to be More professional. You want it to be more casual. This is something that's going to help you. Now, a really underestimated thing about Gemini that I found that you can really do is you can do decent, decent research. So I'm going to ask it. Who is Demis Hassabis? Okay. And Demis Hassabis is the CEO and co founder of DeepMind. That's been acquired in 2014. But what you can do that's really, really cool about this is that when you ask Gemini information, a lot of people are worried about hallucinations. So if you're worried about hallucinations, which are essentially just mistakes that AI systems tend to make, what you can do is you can double check the response. So you click this little G button right here. And then what it's going to do right now, it's going to evaluate the statements and then it's going to validate that against search results. So with this, if you're trying to use this for research, maybe you're using it for your studies, you maybe using it for, I don't know, whatever it is you're using it for, to write an essay, you can ask it something. And then if you're like, I don't know if that's true, you can see where these sources are coming from. And then you can say, okay, this is good. Rate this, check. And you can say, this is helpful, not helpful. You can see where this is coming from. This is from the UK's official website. This is from Wikipedia. And this piece of information was from this other website. I find that this is really good because when I'm trying to find sources about things that I want to verify, I can go on those websites and, and independently check whether or not the information is true or false. In addition, this is very useful for making sure that there are no hallucinations. Now, what you can also do about this is you can also, with any prompt is you can click this button right here called Show Drafts, and you can check whether or not the other three drafts are going to be better or worse. You can see that these drafts are kind of different in how they answer different things. This one talks about career, education and research. This one is just a paragraph, and this one is just a very simple, like Cliff Notes version of who Demis Hassabis is. So whichever way you decide to do this, I think it's always important if you are confused to use the different drafts because it's actually allowing you to just explore different ways to get a response. Now, one thing that you might actually know if you are in America and you're looking at this screen, you might realize that this screen does look a little bit different. For example, there are different versions of Google Gemini based on where you are. So one of the things I'm going to do now is I'm going to turn on my VPN and I will leave a link to the VPN I use and I'm going to show you how there are some small different abilities that you can do in Gemini if you are in other locations. I'm just going to change my VPN to the usa and then I'm going to show you guys what my screen is going to look like, because I'm sure some of you from the USA are probably wondering why this looks just a little bit different, if you haven't noticed already. So now that I've switched over to a vpn, you can see I'm not sure if it's going to show my location, but currently it says that I'm in California, usa. Spoilers I'm not. But essentially right here you can see that it's unlocked one ability, which is the ability to generate images. And I'm guessing it's using Google's Image in version 2, but you can see that this version is pretty, pretty impressive. So what we can do is we can click that, generate a few stick futuristic car driving through an old mountain road surrounded by nature. And this is something that you can't actually do if you're currently living in the EU or UK for whatever privacy reasons or region restrictions. You can use a vpn. And this is going to be something that you're quickly able to do. You can see it generates these images quite like Dall? E used to do, or quite like midjourney or whatever kind of, you know, system you used to use. But these are the images that you can generate. I would find that when I'm generating these images, what you can also do, if you don't like them, you can click generate more and then you can see exactly how things are. What I've tended to realize about the responses from Gemini is that they are quite good for basic things. So, for example, if I was to say, do a picture of a dog's birthday party, this is something that would be done a lot better than futuristic images. I'm not sure why it is. I'm guessing that maybe Google didn't train on those things, but for whatever reason, this seems to be a lot easier. Like these images you can see right here, they're a lot higher Quality. And they just seem to be a lot better in terms of coherence rather than these other images. Because this one looks like photorealistic. But if we, if I just click off this image and then we compare it to this one, you can see that this one, it looks pretty good. I'm not going to lie. But compared to the photorealism of other ones, it doesn't look as good. Like right here. I don't know what's going on with these wheels. Very, very strange looking Bugatti. But the point is, is that you can see right here, my location, California, usa. This is something that you can do. And if you're wondering why on earth have I never been able to generate pictures before, this is what you can do in this version. Now, another thing that you can use Google Gemini to do is you can actually use it for Google Docs, okay? And this is something that a lot of people don't know because I'm not sure what Google did, but they kind of did mess up when they put the link down. What Google did was they put the link in a very obscure position. But I will put the link down in the description, because when you come to Google Docs, you likely won't see these unique features here. But essentially what you can do is you can use this to write many different things. Okay? So if you don't have this when you go onto Google Docs, okay, which most of you probably won't, what you want to do is, is you want to click the link in the description for Workspace Labs, and then once you click that link, it should say, you're in. And I'm guessing that you do need Gemini Advanced to have access to this. So I would say if you're on the Advanced tab, click the link in the description and then it's going to say, success, you're in. And then you're going to be able to do these features. Basically what you can do is you can just say, help me, right? So I can say click, help me right? Write me a video script about why AGI will change society. And then you can see I can literally click here, create. And I find this a lot more effective than just writing in Gemini because. Because it's natively a Word document. And you can see right here, this looks really cool. And it allows me these different buttons to be able to refine my prompt and be able to do that. So you can see right here, I can insert this. And then, for example here there's always going to be this button. So I can highlight this and I can say here, and then I can say elaborate on this. So, for example, if I wanted to elaborate just to make this longer, and this is a conclusion, which is really nice. So I can see it's doing its thing and then you can see right here, it has all of this information. And then I can literally just click insert. So this is how people are writing documents really, really quickly. Just highlighting this bit, I'm going to click, I can add a different prompt, I can just highlight this, then I can click elaborate. I can shorten different things. And I find that this is really, really effective for writing those longer, more comprehensive documents that you really do want. And I can click insert and you can see in literally just around two minutes in Google Docs, rather than copy and pasting and moving to words and whatever, this is something that is so, so effective. That really, really does help me quite a lot in my day to day. Now, one of the last things that you can do with Gemini is you can create code and you can actually run the code in your browser. So you can see right here it says, give me an example of dictionary comprehensive in Python with examples. So I'm going to click that and then you can see right here. Remember, this is Gemini Advanced only. And then what you can see here, that's really, really cool. I really, really like this, is the fact that you can run these different versions of code to test if they work. Now this is some pretty, pretty basic stuff, but you can always click the Play button right here and you can see what it's able to return. What's also cool about this, what I really do like about Google Gemini is that you're able to literally edit the code however you see fit. Because sometimes what you want to be able to do is you want to be able to, you know, edit it for whatever, whatever you want, whatever you want. I don't know if you, you know, want to do that. You can literally just edit the code and then you can literally just copy that and then move it to a different platform. This is something that I use for very niche things, but it's something that I find really useful that you can actually edit the code because sometimes it does do a mistake, and usually if there are mistakes, you can literally just ask Gemini to once again revert that mistake and just say, look, this is the mistake, please fix this and it will likely update that. Now, if this did help you out completely with Gemini Advanced, if there was anything I did miss, don't forget to leave a comment down below. But hopefully this did help you, because a lot of things were confusing before this.
