Google Gemini: Multimodality in Your Google Ecosystem

Reading time: approx. 10 min

After exploring OpenAI's ChatGPT and Anthropic's Claude, it is time to take a closer look at Google Gemini. Gemini is Google's own family of AI models, designed from the ground up to be multimodal. This means they can not only understand and generate text, but also images, audio, video and code. For teachers who already use Google services like Google Workspace for Education, Gemini offers seamless integration and powerful features to improve both productivity and learning.

What you will learn

• What multimodality means for the Gemini models and its benefits.
• Which Gemini models are relevant for teachers (e.g. Gemini 2.5 Flash and Gemini 2.5 Pro).
• How Gemini can be used for text and image generation and other creative tasks.
• Gemini's ability to handle Swedish and integrations in Google's educational offerings.

The Basics: What is Google Gemini and Multimodality?

Google Gemini is Google's most capable and flexible family of AI models. They are built to understand and combine different types of information such as text, code, audio, image and video. This ability is called multimodality. Unlike older models that primarily focused on text, Gemini can receive an image as input and then respond with relevant text, or generate both text and images in the same response.

Different Gemini Models for Education

Google offers several Gemini models, some of which are particularly relevant for the education sector and available via the Gemini app:

• Gemini 2.5 Flash: This is often the default model in the Gemini app for most users, including students under 18. It is optimized for speed and giving direct answers, which makes it excellent for everyday help. You see it as "Gemini 2.5 Flash" in your interface.
• Gemini 2.5 Pro: This is Google's most capable AI model for complex tasks, available via the Google AI Pro plan or certain Google Workspace for Education add-ons. It excels in reasoning, instruction following, coding and creative collaboration. Gemini 2.5 Pro has a very large context window of 1 million tokens (which can handle up to 1500 pages of text), making it ideal for in-depth research and analysis of large amounts of data. You see it as "Gemini 2.5 Pro" in your interface.
  • Within education accounts you can also get access to Gemini 2.0 Flash Thinking Experimental and Gemini 1.5 Pro with Deep Research, which are optimized for showing thought processes and performing comprehensive research reports respectively.

Strengths: What is Google Gemini Good At?

Gemini's multimodal nature and integration with Google's ecosystem provide unique advantages for teachers:

1. Multimodal Understanding and Generation:

   • Text to image: Create images directly from text descriptions.
   • Image to text/analysis: Upload an image (e.g. a diagram, a handwritten note, an image of a historical object) and ask Gemini to analyze it, describe it or ask questions about it.
   • Image to video: Gemini can transform still images into short video clips with audio. (Unfortunately not yet available in Sweden, but check if it has changed.)
   • Practical example: "Generate an image of a Roman legionary in a modern city" or "Analyze this diagram of the water cycle and explain it at a level for grade 5."

2. In-depth research and analysis: With Gemini 2.5 Pro's large context window you can upload large documents, such as textbooks or research reports (up to 1500 pages), and ask Gemini to summarize them, answer detailed questions or generate practice tests based on the content.

   • Practical example: Upload a chapter about World War II and ask Gemini: "Create a summary of the main causes of the war and generate ten multiple-choice questions based on the text."

3. Integration with Google Workspace: Gemini is deeply integrated into Google Workspace apps like Gmail, Docs, Sheets and Slides. This streamlines administrative tasks and material creation:

   • Gmail: Draft emails.
   • Docs: Write, edit and improve texts.
   • Sheets: Organize data, fill in columns and automate text processing.
   • Slides: Generate presentations from notes.
   • Practical example: In Google Docs you can ask Gemini: "Write a draft lesson plan for a social studies lesson about Sweden's geography for grade 6."

4. Pedagogical focus (Gemini for Education): Google has a specific version, "Gemini for Education", which is built with higher data protection and privacy, where student data is not used to train AI models. It also includes tools for AI literacy and stricter content policies for students under 18.

Swedish Language Handling and Image Generation

• Swedish: The Gemini models are trained on a comprehensive multimodal and multilingual dataset. They have broad language coverage and perform well in Swedish. You can draft, summarize and translate texts in Swedish with good quality.
• Image generation: Yes, Gemini can generate images based on text prompts. You can also edit existing images or ask Gemini to generate both text and images in an interleaved format (e.g. an illustrated recipe book or a story with pictures). Gemini 2.0 Flash can generate images in 1024px, and there is also a new photo-to-video function.

Common Pitfalls and How to Avoid Them

Like other AI models, Gemini has its limitations:

• Hallucinations: Although Gemini is powerful, it can sometimes generate incorrect or misleading information.
  • Solution: Always fact-check and verify information generated by AI, especially when it comes to critical facts or student assessment. Gemini for education accounts includes a fact-checking feature that uses Google Search.
• Data protection and privacy: Despite "Gemini for Education's" improved data protection, it is important to follow the school's policies for personal data. Student data should not be entered into general AI services that do not have specific agreements in place that guarantee data protection according to GDPR.
  • Solution: Use Gemini to create general resources, lesson plans and exercises, not to process individual students' personal or sensitive work.

Implementation in the Classroom

• Time-saving for teachers: Use Gemini to quickly generate drafts for lesson plans, assessment matrices, test questions or parent information.
• Differentiated learning: Ask Gemini to adapt texts or exercises to different reading levels.
• Enhanced research and understanding: Students (with appropriate age and supervision) can use Gemini to summarize complex topics, generate practice tests based on their own notes or get help with step-by-step problem-solving.
• Creative projects: Explore image and video generation to illustrate presentations or create visual stories.

Next Steps

In the next module we will look at Microsoft Copilot, and how this AI assistant is deeply integrated into the Microsoft 365 ecosystem to revolutionize productivity and collaboration in the digital work environment.