
Explore the three flavors of Google Gemini—Nano, Pro, and Ultra—and see why Gemini Pro provides free, versatile multimodal capabilities, with Nano offering low memory and high memory on-device variants.
Navigate two pathways for technical and non coding students, focusing on bard, prompt engineering, google ai studio, and gemini sdk usage to improve prompts for all.
Explore the Google it button, a double-check feature that fact-checks Bard responses by searching Google, highlighting green verified results and orange uncertain or conflicting results with links.
Organize Bard chats by renaming, deleting, pinning, and hiding; share via public links or export to docs, noting that public links reveal text but not uploaded images.
Visualize data with Bard by generating or supplying data sets, creating line, bar, pie, and scatter plots, and exporting to Google Sheets, with matplotlib code and customization tips.
Explore how Bard integrates with Google Workspace, granting Bard access to Google Drive and Gmail to search and summarize across documents, create marketing copy, and export to Google Docs.
Discover how Bard integrates Google search to access real-time data, current news, stock prices, and web images, while noting training cutoffs and the need to verify timeframes.
Google renames Bard to Gemini and launches Gemini Advanced, built on Gemini Ultra with a two-month free trial. Upgrade provides access to Gemini Ultra and two terabytes of cloud storage.
Access to the Gemini Ultra model currently comes through the Gemini Advanced consumer subscription, with the Gemini Ultra API for developers slated for phase two.
Compare how Gemini Pro, Gemini Advanced Ultra, and Gemini Ultra explain math, biology, and chemistry questions with detailed step-by-step reasoning and accurate answers.
Explore how examples in prompts shape outputs by showing the format, pattern, and content you want, with demonstrations on producing a two-color palette with hex codes and sentiment analysis.
Learn multimodal prompting techniques for image inputs using Bard, Gemini APIs, and Google AI Studio, including structuring prompts, describing images first, and directing attention to specific details for accurate results.
Gemini API starts free with a 60 queries per minute limit; a paid pay-as-you-go tier offers higher limits, and input and output data may be used to retrain Gemini.
Discover how Gemini Pro Vision handles multimodal inputs in Google AI Studio and Vertex AI, detailing token limits, image prompts, and tasks like scene description and object identification.
Build structured image prompts with Gemini Pro Vision to extract book titles and authors, format results with a dash, and iteratively refine author-diverse similar-book recommendations.
Explore configuring safety settings in Eye Studio and the API, adjusting harassment, hate speech, sexually explicit, and harmful content thresholds with safety_settings. See how different blocks affect prompts.
Learn how to build Gemini chat bots with the SDK by starting a chat, sending messages, and using built-in history to maintain context given the model has no memory.
Learn to use the Google Generative AI node SDK to send text prompts with Gemini Pro, handle API keys securely, and process asynchronous responses with generate content.
Explore how Gemini returns responses and how to craft prompts using single strings or arrays, accessing result.response text, finish reason, and safety ratings, with practical JavaScript examples.
Configure safety settings by building a safety settings array with harm categories and thresholds, then pass it to the get generative model to block harassment and dangerous content.
Learn to build a chat bot using the node sdk with initial history and guidelines, test prompts in Google AI Studio, and integrate the generated history into a JavaScript interface.
This course has just been updated
Google just launched Gemini, its largest and most capable family of AI models. Learn how to use this groundbreaking GPT-competitor that's reshaping the landscape of AI! Whether you're a tech enthusiast, a budding developer, or an established programmer, this course is your gateway to mastering the cutting-edge Gemini technology, which is completely free to use and build applications with.
What You'll Learn:
Gemini and LLM Basics: We start with a detailed introduction to Gemini. Is it the next GPT killer? We explore its variants - Nano, Pro, Ultra - and learn the fundamentals of Large Language Models (LLMs) like tokens and context.
Google Bard/Gemini Expertise: Learn to harness the power of Google Bard (now called Gemini), powered by Gemini Pro and Gemini Ultra. Master the art of multimodal prompts, and integrate Bard with your daily digital tools for summarizing emails, managing Google Drive documents, visualizing data, writing code, and more.
Prompt Engineering Skills: Delve into the art of prompt engineering, in a no-BS section focused on teaching you the techniques that actually matter. You''ll acquire essential skills in multi-modal prompting to elevate your interactions with AI.
Interactive Generative AI Studio: Get hands-on with Google's Generative AI Studio. Learn to build, test, and iterate on prompts using a user-friendly graphical interface, and discover how to fine-tune, batch run, and adjust model parameters effectively. Google AI Studio is an invaluable tool: it's like a workshop for writing the best possible prompts!
Master Gemini Vision Pro: Learn to use Gemini Vision Pro, designed for image and video inputs. Learn to craft complex prompts for object identification, generate marketing copy from images, and more.
Model Parameter Knowledge: Unravel the intricacies of Gemini's model parameters including temperature, top k, top p, stop sequences, and safety settings, gaining control over your AI experience. Learn when to use each parameter to finely tune your prompts and API calls.
Building Apps with the Python Gemini SDK: Learn to write Python code to interact with Gemini and Gemini vision. Write multimodal prompts, work with streaming responses, and build dynamic apps with Gemini.
Node SDK Proficiency for JavaScript Users: Prefer JavaScript? We've got you covered! Explore the Node SDK for Gemini, offering versatility in development language choice. You'll learn how to interact with Gemini through the Node SDK.
Google's Vertex AI Platform Vertex AI is Google's enterprise-level AI platform built for apps at scale. We'll learn how to use the Gemini models through the vertex platform. From registration to authentication, from using the Visual Vertex AI Prompt Studio to developing with Vertex AI Gemini APIs, I'll guide you through each step.
Who is this course for?
Non-developers eager to integrate AI into their daily lives.
Developers aiming to build innovative AI-powered applications.
Tech enthusiasts curious about the latest advancements in AI.