Getting Started with Gemini 2.0
Key Takeaways
The video demonstrates the capabilities of Gemini 2.0, a multimodal AI tool available in AI Studio, and showcases its features and applications in various tasks such as creating YouTube thumbnails and automating tasks. The tool is compared to other models like 01 Preview and LLaMA 3.37 billion parameter model on Artificial Analysis website.
Full Transcript
what's up folks welcome back to the channel in this video I want to show you the capabilities of Gemini 2.0 flash experimental this recent release by Google that offers this multimodal live option with Gemini and in the playground in AI studio. google.com which is where I am right now you can interact with it and we're going to do that and we're going to try out the three main features of it so the talk to Gemini option the show Gemini option and the share your screen option so that you see the kind of stuff you can do so when I click here on talk to Gemini this is where I get hey how's it going I'm doing a video about how to use AI for doing certain tasks on your computer what kind of tangible tasks do you think are really nice to automate using uh llms and AI in general that's an interesting topic I think that automating tasks such as writing emails summarizing long documents and transcribing audio files are all very useful applications of llm okay sorry sorry this is really interesting but what I really care about is like workflows that involve you know integrating different providers and you know doing something a little bit more complex than just you know summarizing emails and things like that what do you think are like interesting options I understand so you are looking for more complex automations I think that using AI to manage your calendar and scheduling meetings or to create and manage social media posts across different platforms could be interesting options do these ideas resonate with you yeah these ideas are great thank you very much all right folks so as you can see I'm going to refresh this page so we can see the options again but as you can see the talk to Gemini option is super fast you can interrupt and just like in the um uh real time API by open AI you can talk over it and it's um you quite powerful quite interesting right and the second option that I think is quite interesting is is if we click on show Gemini and then use our webcam to talk about different things in our uh put in front of the put in front of the model and see if the model knows about it so I'm going to try that with different objects here and we'll see what happens all right uh so I'm going to click on show Gemini hey can you tell me what you can see sure I can do that I can see you are in a room with white walls and some framed PR hanging behind you I also see you wearing a baseball cap in glasses is there anything else you'd like me to describe yeah what am I holding in front of you you are holding a round white object with the word Loop printed on it would you like me to tell you anything else about it yeah can I try to guess what that is sure I can try to guess what it is would you like to give me a hint or should I just take a shot in the dark uh take a shot in the dark thinking that it's something related related to noise cancelling okay based on what you said I'll guess that the round with loop on it is a noise cancelling earplug am I right yes that's right how about this what is this okay I see you are now holding a small white rectangular object with a silver metallic part it looks like a USB drive is that right yes that's right how about this this a pretty easy one give me the word certainly you're holding a small black action camera is there anything else I can help you with how about this what is this you're now holding a white bowl that appears to be empty except for a few small very good thank you very much all right folks so as you can see not only is fast and responsive but it can do pretty pretty cool stuff you can you know interact with it and potentially help you with I haven't really explored all the cool things that one could do with the camera option but uh as these models are multimodal now uh natively I think that offers a bunch of options that uh will be disclosing in the future maybe I want to do a video specializing on all the cool stuff you can do with multimodo models now the share your screen option folks what I'm going to do is I'm going to first open up pixel Mater Pro which is the uh software I use for creating YouTube thumbnails and what I'm going to do is I'm going to click on uh create create new thumbnail create new document whatever and I'm going to try to see if the model can help me make you know a simple thumbnail so we're going to click over here we're going to come over into YouTube video thumbnail and um we're going to click on I don't know uh let's see we're going to let's just go over to all templates and we're going to go to this default one and we're going to go over here perfect and now let's pretend that I don't know anything about these things in the software here which I don't really know all of them that well let's see if the model can help me out all right uh so I'm going to go on share your screen I'm going to share the screen for my software hey I'm trying to you're going to share your screen to show me your software yeah please proceed when you are ready okay great I'm trying to learn how to use pixel Mater Pro can you tell me about what are these features here on my right like for example this little brush that I'm holding over with the mouse right now sure you are asking about the features on the right side of the pixel Pro interface specifically the brush icon you are hovering over with the mouse that brush icon is used to select and use the paint tool it will allow you to use a brush to paint on your canvas is there anything else you would like to know how about this one about that has these three circles organized as a triangle okay you are now asking about the icon with three circles organized as a triangle that is the distribute objects tool it allows you to evenly out three or more selected objects in your canvas no sorry sorry no that is incorrect I'm holding over the color adjustments feature in pixel Mater Pro can you tell me a little bit about what color adjustments allow me to do my apologies I understand now the icon you are hovering over is the color adjustments tool in pixel Pro this tool lets you modify the color of your image by adjusting various parameters such as Hue saturation brightness contrast and more would you like me to go into more detail about any of those specific color adjustment options uh now I'm trying to U make a YouTube video thumbnail could you help me with how could I start using pixel metor Pro like what should be the first thing that I do okay you're making a YouTube thumbnail and want to know where to start in pixel M Pro first you'll want to add the image you'd like to use for the thumbnail you can do that by either dragging and dropping an image file into the canvas area or by using the file menu and then place would you like help with that process uh yeah I clicked on file and I see like uh new and then new from photos open open recent what is the option I should use okay you clicked on the file menu and see options like new new from photos open and open recent to add an existing image to your canvas you should select the open option from the menu would you like me to elaborate on that process uh no that's fine okay sounds good is there anything else I can assist you with in pixel M Pro all right what can you tell me about the photo that I uploaded okay you want me to tell you about the photo you uploaded however I don't see a photo uploaded on the canvas it's still a white blank canvas did you mean to upload a photo already or do you want help with that process can you look over it now and tell me what you see okay I can look over the screen now however I still see a blank white canvas with no image added to it is there something else I should be seeing uh yeah let me see if I can change that so that you can see the proper the proper image so if I do this uh okay it looks like you've adjusted something but I'm still seeing a completely black screen it's possible there's a screen sharing issue or something else going on yes there is a screen sharing issue I'm going to try to fix that and try again all right folks so as you can see it's not perfect but we're going to try again I'm going to click on share your screen I'm going to click on the uh I'm going to click to go entire screen now folks and I'm going to go over to pixel meter Pro hey can you tell me what you see sure I see that the user is in pixel Pro I can see a photo of a person with glasses and a hat sitting at a table using a laptop there are also some editing tools on the side and perfect perfect sorry I'm working on a YouTube thumbnail and I want to use this photo what should be the first thing that I do the video is about using AI tools to do different things okay so you are creating a YouTube thumbnail with a photo in pixel Pro for a video about using AI tools a good first step for your thumbnail could be to crop the image and adjust the composition so it's focused on the main subject would you like suggestions about any specific part of the thum process yeah uh what do you think I how do you think I should crop this image like what should be the subject the most prominent part of the image is the person using the laptop so it would make sense to crop the image so that the person is the subject of the thumbnail and the surrounding elements are more in the background you can do so by adjusting the cropping handles to focus on the user is there anything else I can help with no that's pretty good thank you very much you're welcome is there anything else I can assist you with no it's fine so folks as you can see it's not perfect because the model sometimes hallucinate things that are in the picture just to give an answer like you just saw with pixel Mater Pro and just inventing functionality but when you give it information about what it is it can be really helpful to na to help you navigate new software and you know work on code and things like that so I still defin recommend you play around with this experiment with it and see what kind of stuff you can do and um to finish things off I think the other thing that you can do with this um uh model is literally write text to it right so you can literally say hey can you tell me about things I could automate using llms like the recent release from Google of Gemini 2.0 experimental sorry not experimental flash experimental and the model is there you gois customer service llm would you like to explore any of those areas in more detail and that's it folks you can interact with it in the Stream real time option we interact with it and the answers are always in text but if I I go into create prompt I can send that to the actual model and get just a text based response like I'm going to do now and as you can see super fast given the cap capability and performance of the model and if actually we can actually inspect a little bit about those capabilities by going to the actual original post Google Gemini 2.0 uh release post and there you go Gemini 2.0 and it's a super powerful model that they're putting behind all sorts of interesting projects like uh the ones that I'm going to show you now like Mariner uh asra and others and you can check out their capabilities and capabilities of the model by checking out this table here in the release post where they show the difference in performance for Gemini 2.0 with comparison to previous Gemini versions and another thing you can do is you can go to a website called artificial analysis. and we can check out Gemini 2.0 know flash experimental to see how it Stacks up against current providers like 01 preview and Lama 3.37 billion parameter model which is amazing because it's like free and as you can see Gemini 2.0 flash it's like in third place right now compared to om preview for Quality apparently according to a this analysis which I love this website for comparing stuff and um it's actually in third place for Speed so Lama 3.37 a billion is now super duper fast um and uh price I think Gemini 2.0 let's see I don't know where it stands on price because the Gemini yeah it's an experimental it's an experimental phase right now but yeah you can check out this website to see more interesting stuff about the capabilities and the analysis like this index of quality reasoning and knowledge and so on and so forth so I guess with that folks I'm going to stay here thanks for watching folks don't forget to like And subscribe and see you next time cheers
Original Description
In this video, we dive into Google’s Gemini 2.0 Flash Experimental, a cutting-edge multimodal AI tool available in AI Studio. Be sure to like, subscribe, and stay tuned for more AI adventures! Cheers!.
📚 Chapters:
00:00 - Introduction to Gemini 2.0
00:14 - Overview of Gemini's Features
00:30 - Exploring the Talk to Gemini Feature
01:57 - Demonstrating the Show Gemini Feature with Objects
04:17 - Testing the Share Your Screen Feature to Help with Creating a Youtube Thumbnail
09:00 - Troubleshooting Screen Sharing Issues
11:06 - Text-Based Interactions with Gemini
12:21 - Highlighting Gemini 2.0's Capabilities and Comparisons
14:00 - Conclusion and Call to Action
🔗 Links:
- Subscribe!: https://www.youtube.com/channel/UCu8WF59Scx9f3H1N_FgZUwQ
- Tiktok: https://www.tiktok.com/@enkrateialucca?lang=en
- Twitter: https://twitter.com/LucasEnkrateia
- LinkedIn: https://www.linkedin.com/in/lucas-soares-969044167/
- Check out my prompt engineering course: https://automatalearninglab.thinkific.com/courses/prompt-engineering-basics
Support the Channel!
- Buy me a cup of coffee: https://tr.ee/7tYsD-tUu2
- Paypal: https://paypal.me/lucasenkrateia?country.x=PT&locale.x=pt_PT
- Buy my prompt engineering course: https://automatalearninglab.thinkific.com/courses/prompt-engineering-basics
Watch on YouTube ↗
(saves to browser)
Sign in to unlock AI tutor explanation · ⚡30
Playlist
Uploads from Automata Learning Lab · Automata Learning Lab · 0 of 60
← Previous
Next →
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A Quick Tutorial on NLP Basics
Automata Learning Lab
Automating your Digital Morning Routine with Python
Automata Learning Lab
Exploring Problem Solving with Python and Jupyter Notebook #1
Automata Learning Lab
Summarize Papers with Python and GPT-3
Automata Learning Lab
An Experiment Tracking Tutorial with Mlflow and Keras
Automata Learning Lab
Automating Google Forms Submissions with Python
Automata Learning Lab
Productivity Tracking With Python and the Notion API
Automata Learning Lab
When your Machine Learning Model Fails Do This ;p
Automata Learning Lab
Machine Learning Tip#1 Practical Deep Learning Course
Automata Learning Lab
Machine Learning Tips: Deep Learning Monitor
Automata Learning Lab
Machine Learning Tips#5 MLOPs specialization in Coursera #machinelearning
Automata Learning Lab
Automatically Changing Desktop Wallpaper with Python and the Nasa Image API
Automata Learning Lab
Building an Image Classifier to Filter Out Unused Images From Your Photo Album with Machine Learning
Automata Learning Lab
Automating VS Code Snippets with Python
Automata Learning Lab
How to Set Up a Machine Learning Environment with Conda and Pip-Tools
Automata Learning Lab
9 Google Search Tips for Machine Learning
Automata Learning Lab
Thinking Tools
Automata Learning Lab
Automating Car Search with Python and Data Science
Automata Learning Lab
Generating Images from Text with Stable Diffusion and Hugging Face
Automata Learning Lab
A Practical Introduction to Data Science using the Spaceship Titanic Dataset from Kaggle
Automata Learning Lab
Jiu Jitsu App with Python and Streamlit
Automata Learning Lab
2 Apps for Coding In The Ipad Pro
Automata Learning Lab
From Tensorflow to Pytorch?
Automata Learning Lab
Building an Audio Transcription App with OpenAI Whisper and Streamlit
Automata Learning Lab
Productivity Tracking with Python Short Summary
Automata Learning Lab
Automating Expense Reports with Python
Automata Learning Lab
ChatGPT, Angry Pandas and AI Code
Automata Learning Lab
7 Strategies To Learn Anything Using ChatGPT
Automata Learning Lab
Building a Thought Summarization App with Whisper and GPT3
Automata Learning Lab
Visualize a Neural Net Learning Polynomial Functions
Automata Learning Lab
Automating Notion with Python
Automata Learning Lab
Pose Tracking for Jiu Jitsu - Update #jiujitsu #machinelearning
Automata Learning Lab
Update to my Pose Tracking for Jiu Jitsu Project #machinelearning #jiujitsu #ai #deeplearning
Automata Learning Lab
ChatGPT API Released by OpenAI
Automata Learning Lab
ChatGPT API Response Format #machinelearning #ai #datascience
Automata Learning Lab
Beyond Stable Diffusion with Composer | Automata Learning Lab Paper Series #1
Automata Learning Lab
Beyond Diffusion Models with Composer #machinelearning #ai
Automata Learning Lab
Machine Learning for Jiu Jitsu
Automata Learning Lab
Prompt Engineering Basics #machinelearning #gpt4 #chatgpt
Automata Learning Lab
Visual ChatGPT: Integrating Images with ChatGPT Paper Series#2
Automata Learning Lab
Visual ChatGPT #machinelearning #ai #artificialintelligence
Automata Learning Lab
LERF - Language Embeddings + NERF for Querying 3D Spaces #machinelearning #ai
Automata Learning Lab
Summarize Papers with Python and ChatGPT
Automata Learning Lab
Large Language Models can use Tools Now! #artificialintelligence #machinelearning #ai
Automata Learning Lab
Sparks of AGI in GPT4? #machinelearning #ai #agi #artificialintelligence
Automata Learning Lab
Toolformer: LLMs can use Tools! #chatgpt #llms #gpt4 #gpt3 #artificialintelligence
Automata Learning Lab
Talking to Your Notes with LangChain #artificialintelligence #llms #gpt4 #chatgpt
Automata Learning Lab
How to Talk to a PDF using LangChain and ChatGPT
Automata Learning Lab
Query Your Own Notes With LangChain
Automata Learning Lab
HuggingGPT #machinelearning #artificialintelligence #huggingface #gpt4 #chatgpt
Automata Learning Lab
Do as I Can Not as I Say Paper #artificialintelligence #llms #reinforcementlearning
Automata Learning Lab
Automating Anki Flashcards with OpenAI and GPT-4
Automata Learning Lab
Building A PDF Summarization App with Gradio and LangChain
Automata Learning Lab
Auto-GPT #artificialintelligence #gpt4 #llms #autogpt
Automata Learning Lab
DocGPT - Chat with Github #artificialintelligence #gpt4 #chatgpt
Automata Learning Lab
LLMs for Research and Planning #artificialintelligence #gpt4 #llms
Automata Learning Lab
How I Use ChatGPT for Interactive Language Learning
Automata Learning Lab
Building an Audio Transcription App with Gradio and Whisper
Automata Learning Lab
Summarizing and Querying Multiple Papers with LangChain
Automata Learning Lab
Mojo - The New AI Programming Language?
Automata Learning Lab
More on: LLM Foundations
View skill →Related Reads
Chapters (9)
Introduction to Gemini 2.0
0:14
Overview of Gemini's Features
0:30
Exploring the Talk to Gemini Feature
1:57
Demonstrating the Show Gemini Feature with Objects
4:17
Testing the Share Your Screen Feature to Help with Creating a Youtube Thumbnai
9:00
Troubleshooting Screen Sharing Issues
11:06
Text-Based Interactions with Gemini
12:21
Highlighting Gemini 2.0's Capabilities and Comparisons
14:00
Conclusion and Call to Action
🎓
Tutor Explanation
DeepCamp AI