If you are a big fan of Google Gemini, you are probably only using 50% of real power. I have been testing Gemini models on Google AI Studio, and the results are impressive. So in this video, I'll show you why you can't afford to ignore Google AI Studio if you are mainly using germ in your flow, let's go. The Gemini web app is what most of us know. It's the consumer friendly chat interface like ChatGPT designed for everyday questions, quick summaries and general assistance. It's simple and effective for most work tasks. But AI Studio on the other hand, is a free web-based platform designed as a testing ground for power user is where you can assess so much more advanced capabilities from Gemini, with the four core modules chat here you can experiment with the latest Gemini models with 1 million token context window and fine tune settings. Stream allows you for real time interactions with Gemini, Generate Media, where you create images, videos, and audio, all using the latest Gemini models like VEO, Imagen and build where you can deploy production ready apps with Gemini. For this video, we'll be focusing on the first three, as you can use them right away without any coding skills. Perhaps one really big difference I found is the output speed and response quality. The Gemini on AI Studio consistently generate higher quality and more comprehensive output. Like this example for research analysis task for a bunch of super long PDF reports, the output on AI Studio provides more extensive evidence, more granular statistic and detailed breakdown than what you get on the standard web application for the same prompt. And not just that, some of my favorite unique capabilities that are only available on AI Studio: System Prompt, you can use it to define any specific style and instruction to ensure the output quality. Temperature settings. It controls how creative or predictable the Gemini response are. The higher the number, the more the randomness and less predictable. You'll be surprised how useful this is in a minute. Compare Mode. Let's just see how different Gemini models response to the same problem side by side. Our first workflow is one of the single most powerful thing I've found in AI Studio is to get side by side, multi persona perspective from Gemini. With this compare prompt for the same analysis. For example, I'm analyzing a company's annual report with over few hundred pages to identify key strategic insights and action. So first, give Gemini this super long report, and now click this little icon, compare mode, which brings you a side by side chat window. So on the left we can define our system prompt and ask Gemini to add as a skeptical financial analyst. So uncheck sync, which means we don't want to sync the same system prompt for the right side chat and then set the temperature level to 0.2, which means we prefer more precise, factual, low risk answer. And on the right hand side, ask it to be an innovative marketing strategist with temperatures set to 1.5 for more creative and expansive output. Now ask it to provide strategic insights and suggest actions. So this is a super cool way to compare the output from Gemini using different perspective. And not only, I found the output is more detailed. You can compare them side by side. So for example, I noticed the financial analyst says the diversification strategy might be less effective while the marketing strategists frame Asia as a growth engine for the company with data support. So this is contradicting views and it's exactly what we want to get a diverse perspective on the same issue and inspire your own thinking. And don't forget to save the chat if you like the response or to save your time. You can just change to the default setting to auto save. And not just for defining persona. Maybe you just want to do AB testing for content creation, using different models, system prompts. You can apply the same concept and compare the output side by side, making the process much more efficient without going back and forth. And if you're excited about the potential of Gemini for work, especially for marketing, I recommend this free ebook from HubSpot, "Google Gemini at Work". You can find it down the link in the description. This guide shows you how to leverage the five-tool Gemini marketing stack that includes deep research, NotebookLM, Gemini 2.5 Pro , image generation and learn about and how they can work together in different common marketing use cases like research strategy content, plus a practical four week implementation plan with Gemini. So you are using Gemini beyond just an AI chat bot and streamline your marketing workflow. What I found useful is how to provide the ready to use prompt templates, like "The Complete Strategy Template" and "The Content Multiplier Template". These give you great starting point. You can copy, paste, or customize for your own marketing needs so you don't have to start from scratch. Download this in the description below for free and thank you HubSpot for sponsoring this video. Okay, you might have done incredible analysis but like most people, presenting it confidently is a whole different skill. Now you have to present it to your stakeholders or teammates or clients. And that's where another incredible workflow comes in that is to ask Gemini to be our live presentation coach and get real time feedback. For example, I am going to present this marketing analysis to my stakeholders and now on AI Studio. This time go to the stream, which we can interact with Gemini in real time. We can first define the role and responsibility for this presentation coach using the system prompt, with areas to look for, and how it should response. And then pick an output voice that you like, and then click share screen for the presentation. And you can also pick the talk and webcam mode, depending on how you want to interact with Gerini, and it will immediately bring up your presentation. Good morning everyone. Um, so the childcare market is like really big right now with value focused parents who I'm, you know, better our eye on education. So we should better position ourself to capture this opportunity. Okay, I'm done. Any feedback for my presentation? Let's refine that opening. Cut the filler words like um so and like to sound more confident. Also, instead of saying we should better position ourselves, try a more direct statement like we must strategically position ourselves to capture this market opportunity. Okay, then for this slide, how should I present to make sure it is engaging for my audience? To engage your audience, try focusing on the key benefits each segment represents. Use specific examples or maybe tell a brief story about how these changes are impacting families or educators. This will make the information more relatable and memorable. So you just see that. I think this is one of the best ways to use Gemini on AI Studio. So basically you can just have it to give you feedback on any kinds of important presentations, meeting perhaps for immediate improvement. And the setup is so easy. Just put the system prompt on what you want it to give you the feedback, and then share your screen. And not just for presenting, sometimes if you’re just get stuck in the process, like not knowing how to use certain software or tools, or want to get quick answers or ideas, you can just share your screen and ask Gemini to trouble shoot for you. So there are a lot of ways you can use this powerful stream realtime capability. So you just see the live feedback from Gemini. But now let's explore Gemini's creative capabilities on AI Studio. So you can actually create and edit visual content at a professional level, something that traditionally requires expensive softwares or design skills. And let me show you what I mean with real-world examples. So I have these two images, a beach towel scene image, and a backpack product image. And I want to merge them together to create a product moodshot. So on AI studio, go over to generate media and pick Gemini image generation. So here we'll be using the flash model for image generation, which allow us to combine multiple images natively. Give it the two images and keep all the default settings and then prompt it to edit the backpack product to put in the scene and make sure it appears as a hero product and the output is amazing. So Gemini has kept the same texture as in the original product image. It looks consistent and it's so detailed for the backpack product in this moodshot, and it blends super well with the scene, looks very natural, and to make it as the key focus with bright lighting as we instructed it. And this is only my first attempt and Gemini got it right. I also try the same prompt on the Gemini web app, but I must say that the output is definitely better on AI studio. But we are not done yet. So this time we have this static infographic for social media. So we want to turn it into an animated video. So again, on generate media. So this time click VEO, which is the latest state-of-art, Google video generation model. So for the settings, we can keep it to the default and give it the static infographic, and then ask it to turn it into an animated video with movements added to the illustration that is suitable for business, social media. And immediately, we have an animated infographic, and I like these animated illustrations making it so much more engaging and you can turn any of your static social images, infographics into animation using this method. Besides image to image, or image to video, you can always use the latest Gemini Imagen model on AI studio to generate entirely new images for free using just a text prompt, the output quality is so high, they are ready to be used in your professional projects immediately. But note the video generation has quite a strict usage limit for its free tier. So spend some time to craft your prom before you run it. Another bonus tip is try to minimize the text or elements in your original static image, because for now, it's still a challenge for LLM to render the text correctly in video. But this can be applied in so many ways. Animated product moodshots. Animated social graphics, portfolios, presentations. So basically it will make the experience much more engaging. But here is what makes Gemini even more powerful on AI Studio. It doesn't just create content. It gives you incredibly detailed, granular output that breaks down every single elements. So the coming workflow solves one of the biggest painpoint every business face: turning training videos or process tutorials into actual documentation that your teammates, your clients can follow step by step. It can be YouTube videos, your own recording raw file, or loom screen recording. Gemini can do the work for you. For example, I want to turn this existing YouTube tutorials about Google Analytics 4 into a set of ready to use process documents, so I can share that with my team along with this video. So on AI Studio, you can import any YouTube videos and you can even specify the timestamp, which is powerful. And you can import more than one YouTube videos as well. And it doesn't have to be YouTube. You can also just upload your raw screen recordings for any process, and then just ask it to create a step by step process that we can use. And then you can see the generated process is so detailed with the key sections. Each of the key step and instruction, and I found every time the response I got from AI Studio is much better than on the Gemini web app web app and you can just copy and paste to a work document that is ready to share along with the video. Saving you much time on preparing those tedious processes. And this is also something I'm increasingly using it for myself, and we can even ask it to create a two minute summary script, which we can use to generate an audio for internal team briefing, and then go to generate media for Gemini Speech Generation past the script. And then one click to generate the audio track. So you can do all this within one single platform. So I recommend you check this out. Just try a simple screen recording and the output is amazing. Now what if, instead of just process instructions, you need to prepare for important conversations, job interviews, client meetings, strategy discussion. AI Studio can generate full podcast style dialogues with multiple speakers having actual back and forth conversations powered by Gemini Native Audio tech. So unlike the training audio we just created, it is about crafting realistic scenarios and let it become your ultimate preparation tool. For example, I'm preparing for a job interview for marketing director role at this tech company, and this is a real job ad I grabbed from LinkedIn. So now on AI Studio, give it both the job ad and my resume and then ask Gemini to be the interviewer, Sarah, and create a mock interview with me and to ask question based on the job description and my experience and to make sure there are behavioral question, because these are super popular and common for interviews, especially for leadership position. And you can see the output script is great. It is in a dialogue format with lots of great details. And I really like this question about how I particularly envision, leveraging AI to enhance marketing function, which is super relevant, strategic, and targeted for my experience and the role. And you can see all the responses drafted by Gemini are thoughtful as well. Or you can even give it to other model like Claude to get a different opinion. Now we can just copy the text and generate the audio using the text to speech functions powered by Gemini, together with the style instruction. And make sure you specify the speaking name in the settings. So give it a few minutes to generate the dialogue Audio. I'm Sarah, the CMO here at Safe Software. I've had a chance to go through your resume in detail, and I must say, it's quite an impressive and comprehensive background So this will significantly help you to prepare for any important meetings and you can just download it for your own practice. Note for now. The chat history is not applicable to the TTS audio generation, and sometimes I found generated audio might be cut off if it's too long, but still, this is a super impressive workflow. AI studio is not without downsides besides the usage limit set on the free tier. Google might use your data to improve their products whenever you use what they call the unpaid service, including AI Studio, even if you access it under an organization account. So if you are privacy conscious, consider using the Gemini API on the platform or on Vertex AI with billing enabled. And if you want all the prompts I share in this video and other feature content, join my community. I will put the link below. And before you go, also watch this video about how NotebookLM and Gemini can work together to unlock some more powerful workflow. I will see you next time.