The best Google Gemini prompts you haven’t tried yet

If you’re just getting started with Google’s Gemini, you may not realize how much it can do. It’s more knowledgeable than Google Assistant, which often has a hard time answering even basic questions. And thanks to recent advances in generative AI, Gemini can answer any question under the sun. With such endless possibilities at your disposal, let’s take a look at some of the best Gemini prompts you can use to get the most value out of your chatbot.

Google Translate has served most of us for over a decade, but it doesn’t always do a perfect job. Case in point: While traveling in Southeast Asia, I often had trouble translating short texts using Google Translate. Single words on restaurant menus were the most problematic because the app couldn’t understand the context of the words. The same word can mean dandruff or a piece of meat depending on where and how it is used.

Gemini doesn’t suffer from the same pitfalls, as its training process requires scanning hundreds of gigabytes of text in different languages. This allows you to understand nuance and context. Additionally, Google’s past technical documentation shows that this language model often outperforms Google Translate in standardized tests and benchmarks. So, next time you need to translate something in a complex foreign language like Japanese, try Google’s chatbot.

The Gemini prompt is also very simple for this use case. Just type “Translate the following text into English” and copy the text you need translated.

Did you know you can ask Gemini to create AI-generated art for you for free? This feature is clearly not perfect, as it recently came under fire for producing overly diverse images. However, it still works well if you are generating fictional or artistic images and don’t care about historical accuracy. This is certainly a useful alternative to many other AI image generators that require some form of payment.

To have Gemini create an AI-generated image, send it a prompt that starts with “Create an image of…” followed by a vibrant text description of the art you want to create. The more details you include, the better the results will be.

When planning a vacation, I usually use multiple Google services like Flights and Maps. Then, after choosing the best option and route, proceed to create your itinerary in a spreadsheet. But thanks to Gemini Extensions, you no longer need to perform these steps individually or manually. Chatbots can interact with various Google services all by themselves.

First, you need to enable the Gemini extension. setting > Extensions menu. To plan your trip, choose the Google Flights, Hotels, Workspaces, and Maps extensions.

Finally, enter a prompt like “Find the cheapest one-way ticket to Tokyo in May.” Google already knows your location, and it did just that in the screenshot above. Another Gemini prompt is: “Plan his 5-day trip to Tokyo by suggesting flights and hotels under $200.”

Do you prefer reading to watching videos? I do too, and I’ve found that Gemini can be a valuable tool to bridge that gap. When you come across a long video, copy its URL into a new Gemini prompt and ask the chatbot to create a summary. Or ask questions based on the content of the video, like in the screenshot above. This feature is useful when you have watched a video once and need to refer to it again to get just a little bit of information.

When you send a YouTube link, Gemini doesn’t actually watch the entire video. Instead, read the transcript attached to the video and use it to summarize the text or as a reference to answer the questions. Most English content has auto-generated or user-uploaded captions, so Gemini will likely handle the majority of YouTube content out there.

Are you having trouble interpreting abstract art or need suggestions on how to redecorate your living space? In recent weeks, I’ve asked Gemini to perform both of these tasks, and The results were amazing. As you can see in the screenshot above, Gemini correctly guessed that his four framed art pieces in my photo represented the changing of the seasons.

Gemini’s underlying language model is multimodal. This means it can interpret non-textual content such as images and videos. The only other mainstream language model with vision-like features is GPT-4, but you have to pay for ChatGPT Plus to access it. Gemini, on the other hand, allows both free and paid tier users to upload images.

To upload an image along with your text prompt, simply click the “Upload Image” icon embedded within the chat box. Then, with the Gemini prompt, ask a follow-up question like, “What is this piece of art trying to convey?” “How can I redecorate this space without spending too much money on new furniture?”

Google has significantly enhanced code literacy since the release of Gemini. In fact, with a recent update, you can now run Python code directly within the chatbot’s interface. A Gemini Advanced subscription is required, but without a subscription anyone can create templates, debug errors, and even translate a few lines of code from one programming language to another.

When comparing Gemini and ChatGPT, we found that Google’s chatbot was able to provide code that worked as reliably as its competitors. Additionally, Gemini can search the internet for up-to-date information, whereas ChatGPT’s free tier limits knowledge. This difference alone can help chatbots generate more accurate code or find up-to-date solutions to errors.