NEW AI Tools For 2025 (Must Know!)
I've dived into hundreds of the latest AI tools and updates and in this video I'm going to show you which ones you shouldn't Miss and one of my favorite AI video updates is the ability to add AI video elements into liveaction recordings so I've got myself a lovely assistant here we've got everything from capable voice agents that can take calls on your behalf that's exactly right I can take calls on your behalf two completely free highquality AI video generators as well as a whole lot more the end of this video you'll not only know some of the latest exciting AI tools available but you'll understand how to use them first up we have Quin 2.5 and this is a large language model set that allows us to also create high quality images and video and at the moment it's entirely free let's take a look at how it works so you can come to Quin sign up and you'll have the option to choose from any of the available models I recommend that the best of the moment is Quin 2.5 Max now you can create videos directly inside of this simply by going to the video generation Tab and as you can see it allows you to enter impr prompts and get incredibly realistic AI videos all for free here's a one of a huge elephant leisurely drinking water by the clear River here is a video that I rendered myself and as you can see that it's doing a very good job at keeping anatomy in the right place and rendering out perfect human hairs let's take a closer look at some of the examples coming out of quen now I ran these problems through both quen and clling to take a look at how these two compare now this is the Quin output and you can see we get a lovely sense of detail in this woman and even the soft fine hairs on her cheek are rendered perfectly I particularly like that they've added in at least one mole here for some slight imperfections now comparatively the cling output is quite shocking she is perhaps overly detailed and it becomes almost disconcerting there is something slightly uncanny about this version now I've put together a whole pack of 10 different comparisons from quen and clling I've also included the prompts I've used and so you can take a look through these at your own Ledger if you like I'll leave a link to those in the description below if you're new here I'm AI Samson and on this channel we explore the creative potential of AI but this is one of the most impressive examples coming out of and you can see it handles the challenging Reflections absolutely remarkably well now if we look at the same prompt from cling you can see that although we have a good output the Aesthetics are leaving much to be desired in comparison now let's look at one of the simple yet unbelievably challenging elements for a video model to produce and that's a person walking now this example comes directly from cing and you can see that it handles the person walking remarkably well that she's walking in an anatomically correct way and there is nothing that belies our sense of reality next up we have the quen output and here we have a slightly overly airbrushed aesthetic and her eyes seem as if they are almost robotic I would have to say but there is a lovely sense of depth of field in this shot I would have to give that one to cling now it's not just video that we can generate all inside of Quin we can also generate images now we're going to compare how these images come out to a number of other AI models later in the video now what is particularly interesting about this model is being able to work with all of these different types of media directly inside of the chat so you can start to discuss the media and the model understands exactly what you're talking about for example I can ask it to change this image to black and white now what it does is it reruns the image with a new prompt however you can see it's not entirely black and white so you might say something like make it more black and white now quen has a number of different features including artifacts which allows you to have a concurrent window open to run code that you write inside of quen it also has the ability to search the web and overall it's an incredibly impressive model now agents are exploding across the world right now and they are certainly the hot topic for 2025 and one agentic option that I want to display to you is is the remarkable things that are going on over at 11 Labs now 11 Labs voice agents are conversational AI tools capable of managing customer support scheduling and interviewing these agents can make and take calls on your behalf unlocking a range of possibilities for Automation and efficiency now I would use this to handle uh my own challenges with keeping up with customer support if you have any business where you are handling customer inquiries you can have a voice agent that is able to ring them up and give them support as and when they need completely trained on your existing knowledge base so you can easily go onto the 11lbs website and test out their existing voice agents to see what they're capable of now they have a number of different situations for voice agents and a variety of voices that you can play around with now you can also have the voice agents pretend to be a character so this can be perfect for integrating as an NPC into video game development and this is so exciting another seeker of knowledge finds their way to my humble archives do you seek knowledge of ancient artifacts forgotten prophecies or perhaps something even more adventurous oh I want to go even more adventurous young lady what have you got in store for a young buck like me these lands are filled with Mysteries waiting to be unraveled okay enough galaban thing around the voice agent World here now to create your very own 11 Labs voice agents all you have to do is come to the conversational AI tab go to create an agent and from here you'll get to select the type of agent that you want to create or you get to start from a blank let's go with a support agent and we'll go with create this AI agent next up you get to choose the language and any additional languages that they might be able to speak then it's as simple as adding in a prompt to explain to the agent what type of role that are partaking in and uploading your own knowledge base so this is where you get to add all of the information relevant to your agent this could be commonly asked questions about your products and services or it could be simply a backstory for your agent to use as their own personality finally when you're ready all you have to do is go ahead and press test AI agent let's have a little phone call with them hi Eric I'm curious about joining the AI filmmaker Academy and I wanted to know what pain options do you provide hi there I'm excited to hear you're interested in the AI filmmaker Academy it's a fantastic program we offer a variety of payment options to make it as convenient as possible for our students we accept major credit cards debit cards and even have financing options available would you like me to provide you with more details about the specific payment options or help you find the best fit for your needs now to deploy your agent all you have to do is go ahead and share the link available inside of the platform you'll then be able to view the conversation history as others use your agent there's also the option to create an phone number that people can use to directly contact the agent now depending on which plan you're on you get a different amount of conversational AI minutes now starting off for free you get 15 minutes of conversational AI now if you go to the $5 a month plan you get 50 minutes and this increases as you go through the levels now this really depends on on how long you expect the conversations people to have with your AI are now if you're curious how I created the intro to this video where I combined The Voice agent from 11 Labs with an avatar I'm going to explain how I did that first of all I came to render net and generated myself a attractive spokesperson next up I took this image I went into cling and then I created a looping video with the prompt woman talks expressively to camera now I did this by using the same start and end frame this means that the video will play in a loop consistently The Next Step was to Simply go to lips sync and from here I took the audio recording from my voice agent and uploaded it using local dubing finally I went to generate and that gave me the intro however AI agents like this are showcasing the abilities already present in the AI systems to perform complex tasks that pre previously would only have been possible by a human and understanding how we can leverage this at scale to build systems processes and businesses unlocks an unbelievable amount of potential one particularly Innovative use case of this technology is where the mobile phone company O2 created an AI granny that speaks to scammers who are trying to scam old people and it simply keeps them on the phone for as long as possible without giving them any useful data and this is a quite frankly original Innovative and hilarious way to use this technology a y pleas dear did you um say pastry I'm afraid I'm not quite on the right page no no I'm talking about the Play Store application Play Store it's not P stre my screen seems to um open up application open application oh dear I think I clicked something wrong open up application it seems to um have gone black is that supposed to happen how do I get it back it's actually incredibly realistic and natural sounding and it brings up a lot of interest not only about the possibilities for this technology but also for the ethical implications because obviously this is uh a technology that's trying to be used in a positive way but if you think about it from the opposite direction of when scammers are going to be utilizing Mass expenditure of AI voice agents to scam people then it becomes a particularly worrying scenario next up is Gemini 2.0
and there are a couple of features that you may have missed that have been previewed on the new Gemini models that are absolutely mind-blowing let's take a look at those I've partnered with HubSpot to share their 1,000 plus marketing and productivity prompts this is a free resource that genuinely useful for marketers business owners and content creators here's what's inside over ,000 prompts covering things like SEO marketing social media and much much more practical and actionable insights that you can start applying right away customizable templates to fit your specific business needs whether you're brainstorming campaign ideas looking for new business opportunities or hoping to streamline your workflows and processes this kit can help you I found it particularly useful for staying organized with all of my different projects and video ideas helping me speed up my process from idea Generation all the way through to delivering a video you can download it for free using the link at the top of the description below and a huge Thanks goes out to HubSpot for sponsoring this video so there is this incredible AI agents demo from Gemini 2.0 that showcases a whole host of remarkable features that are quite simply tantalizing these include native audio output native image output native tool use spatial understanding video understanding and multimodal live streaming so let's dive into some of those and take a closer look now the first thing that's very interesting for creatives is that Gemini 2.0 can natively generate images as part of your conversation for example you can take a car like this one and then you can ask the chat to Simply update the image to turn this into a convertible now you can do that in natural language and what's incredible about this is the current work process you would use to do this would involve masking out parts of the car applying difficult prompts that need to accurately describe the scene and can often be hit and miss process as I'm sure you'll imagine you know the process from working with generative fill in Photoshop or working with editing images inside of mid Journey but here you can remarkably see that the exact output has been generated with just one simple prompt and it's kept all of the elements exactly the same inside of the image now one thing I'm particularly interested in for taking this into a video workflow is the ability to Define actions inside of the image for example here the user has circled the handle of the Comm and ask the AI to open this and then in just 13 seconds it has now generated an image with a car door open whilst keeping every other element of the image intact and this is remarkable if we're going to use AI video to create a first frame and an end frame and then ask the AI video model to animate between those spots now the issue with this is that the new output modalities are available to early testers only however they are saying that they expect to release all of these to everybody else in the next month however Gemini 2 is by far the most impressive model that Google has sh shipped to date and some of the things that they are releasing includes powerful new agent capabilities being able not only just to present information but also to go out and take action and they've entitled this project asra and in this demonstration you can see that they showcase the abilities to bring in both multimodal memory as well as realtime information so this means that you can work with different content types like video audio and image and also bringing live information from The Real World what can you tell me about the sculpture the sculpture you're seeing is called my world and your world by Eva Rothchild located in lwis Cubit Park in London this demonstration it shows a man taking a video call with the AI and showing it a sculpture and it's able to understand what the sculpture is where it is and exactly the required information about that piece of sculpture now although we cannot use all of these presented actions we can use the latest models of Gemini for free and there are some pretty cool things we can do for example we can generate images directly inside of the chat using their leading imagine 3 model and we're able to work with images for example we can ask for an image of a golden retriever now the image model used inside of Gemini is incredibly interesting and one that certainly deserves a closer look and this is called imagine three and the reason why I think it deserves a much closer look is because in this leaderboard where images are ranked in a blind Arena test where you are given two images for one prompt and asked without knowing which model they come from which image you prefer and in this ranking the Top Model is Imagine three it outscores recraft idiogram and Luma photon now we can use imagine in a different way and there is a way to use it in image effects and what's great about this is it gives us a more native image generation user experience now imagine 3 performs particularly well at creating surreal scenes like integrating this hummingbird with the strawberry now the photo realism is also particularly impressive just look at the fine pores on this woman generating nature style scenes it is excelling if you look at this wonderful squirrel I want to draw your attention to the beautiful depth of field here and how each of these snowflakes is beautifully crafted in the image it also handles complex lighting situations with ease here in this neon lit situation the protagonist has a beautiful consistency with the lighting sources with the red Shadows being cast on the right side of her face and then the green highlight lights on the other side it also handles text in images remarkably well I'm particularly enamored with this feather written word of light and the colors and the iridescence on the feathers are quite beautiful now you can use this in their image app setup and what this allows us to do is to work with our own prompts so you can start off with a I'm feeling lucky prompt and it will immediately go ahead and generate these now we have the option of generating IM Imes in different aspect ratios and they generate at an incredibly quick speed but I went ahead and I took an image from the following prompt and it was a Vintage Film photograph of a young Korean woman and I wanted to look at this and compare how it came out in a number of different models so first of all I popped it into imagine and as you can see it does an unbelievably good job at creating photorealistic images with consistency and even perfect hands I love particularly how this little drop of icing is just about to fall off the dut now I also went back and popped this into the Quin 2.5 model and certainly looking at this it is slightly more airbrushed and doesn't give us quite the same sense of realism finally I popped it into mid journey and mid journey is my go-to model still and I must say that the aesthetic coming out of mid Journey Bears a stronger cinematic resemblance however the the hand here looks a little bit unusual there's nothing I can distinctly pull out that makes it feel unrealistic but there is something I don't trust about this finger here it's almost being inserted into the donut now if we were to zoom into the protagonist of each of these three and look at them in one go it becomes evident that quen is the weakest model for image generation and then you would say for realism mid journey and imagine are on par with the key difference being here uh your personal preference and taste and for me I would say that aesthetically I prefer the more muted tones coming out of mid Journey so talking of mid Journey it's worthwhile to explore some of their latest feature releases because I think they are incredibly useful and mid Journey are also teasing us with a few exciting things coming up down the road first up it's mid journy mood boards and this is a remarkable feature that allows us to generate collections of images and then generate more images in the style of the images that we have collected let me show you how that works so you come to the personalized section and then what you do is you collect uploaded or generated images to show your target style now you can create as many of these as you like and I've used these in a couple of interesting ways first of all is I created my own icon set perfectly for my brand and for my YouTube channel so whenever I mention let's say aircraft Bing I can have a nice aircraft Emoji popup so you can either upload your images add them from the link or add them from Gallery then once you have your collection when you want to go to create you come into your prompt bar open the settings make sure to turn on personalize and then select your I your special mood board now as you can see I have these little icons of little birds which fit into my style and you can see that they maintain my brand colors and they also maintain other elements that keep a beautiful con assant style now we can go one step further with our own branded icons and we can add an animation to them and to do this simply we have to come to cling AI video go to image 2 video and then first of all you upload a blank empty shot now this is because it's the starting point of the animation then all we do is go ahead and add the end frame where we'll have the icon animated so we go to end and then we'll select our icon pop them in there now you can leave the prompt and then you can go ahead and generate now the great thing about this is that you can play it forwards and then you can also play it backwards so that you both have an animation for entering the shot and also one for leaving the shot let me show you how a couple of these came out here you have a bird and this one I particularly like of the plane as it animates all of the different elements very effectively excellent job now if you are interested in creating money with AI projects I've got a free ebook available in the description and this has recently been updated and it includes a number of different ideas for creating your own AI products I give you not only the ideas an overview of the process and the tools that you'll need to complete the project and you can download that for free in the description below now I've also been working on a style for generating b-roll for my videos and this allows me to create images with a consistent feel so I've been looking to create these almost uh graphic novel deep Illustrated images and I'm able to now recreate images that consistently keep this aesthetic and it's incredibly useful for having montages that maintain a stylistic cohesiveness now this is great if you're working on anything that requires a lot of images if you're generating a series of children's illustrations if you're looking to create products that have a consistent style and if you're looking to make brand work this is absolutely remarkable if you want to have brand images for your business or for your product that maintain a beautiful sense of commitment to your identity this is remarkable now another interesting feature that mid Journey has released is called Patchwork and this is like an infinite canvas where you can map out a lot of different elements in one place it allows you to utilize the mood bold feature in a more visual playful way now the most exciting thing about mid journey is the rumors are that mid Journey version 7 is coming in the next month and that brings a whole host of new potential to the mid Journey platform so some of the unconfirmed rumors around the new MID Journey features include a video model coming which I am absolutely so excited for I cannot wait to get my hands on that also there is a belief that there will be multiple character consistency available inside of mid Journey so you can create multiple characters and have them interact in scenes in complex ways there is also the an expectation for there to be more knowledge and more detail in the new model now another video model that is out now that you might not have heard about is called juanan video so here we have a futuristic soldier holding a gun and the reflections on the materials he is using here gives us a great sense of immersiveness and also the details in the situation are absolutely stunning from the broken window here and these cracked pieces of concrete and then even the composition of the whole entire shot is extremely immersive and beautiful I love this handheld camera feel as well now if we look at the rendering of some natural elements like water I can see that this is absolutely stunning the way that the light is reflecting onto these tree stumps now it's also hand the complex Anatomy per perfectly well here is a woman performing yoga and you can see generally perfect rendering however just notice the hands here that there is some morphing and at some point it looks like she doesn't quite have all of her digits right now just three fingers so you can try it out for free and the paid plan is extremely generous with 150 credits costing $9.99 per month now what is interesting about this huan model is that it allows you to apply specific luras into the prompting process now what this does is is that allows you to take specific models that are particularly perfect for creating certain situations for example there is an anime model that you can use there is also Al a model that allows you to work exclusively inside Monica's apartment from friends which is a pretty unusual model I'm not sure there's a great I'm not sure there's a great number of use cases for this unless you are creating friends parodies then there are some specific ones for close-up macro shots and also one for fashion and this one does a particularly good at this one is particularly good at rendering the volumous illustrious absolutely remarkable beautiful hair now Pico is an AI video model that you may well be aware of however it has performed some quite remarkable updates and introduced a feature that I have not seen in any other AI video model and this is called peer editions now what peer editions does is it's a videoo video model that empowers users to seamlessly integrate any object or character into their existing videos transforming ordinary footage into extraordinary visual narratives so here you can see example of somebody putting in a rabbit into a shot so now you can add incredible elements into your videos and this is an entirely new way to be working with AI video and it's also a wonderful opportunity to blend working in real life with that of AI video and that creates some quite remarkable situations for example here's a man with a monkey being put on his shoulder here is a giant rabbit being placed in the middle of Japan and here you can insert a tiny polar blare into your microwave this is a very interesting application and new opportunity to work with AI video so you can perform this yourself in three simple steps first of all you add a video it could be something you shoot yourself or a favorite clip just to be sure that it's at least 5 Seconds long so you can go ahead and start with a sample video then you can add an image of the element that you would like to include in your video now there are a few options that they give you to start off with might put in this little cute guy finally you add a prompt in your prompt clearly describe how you want your image to integrate into your video and here we have this very friendly little chap appearing inside of this surfing Montage now this is an extremely exciting way to generate viral content and the possibilities are remarkable of creating really engaging effects whilst maintaining a sense of relatability the content that you create now another interesting AI video tool that you might not have seen in use yet is generative extend in Adobe Premier Pro where you can take any type of video and make it longer so if your clip is not long enough you can simply generate more frames with AI and this is again a beautiful example of blending the existing video with AI now if you found this video useful please do me a small favor and share it with a friend that will also find it interesting I'd really appreciate it which tools excite you the most and are there any that I've missed let me know in the comments below I thank you very much for watching I thank you for being here and most of all I wish you a delightful day
2025-02-14 20:06
