What is Google Gemini? How to use Gemini 1.5 Pro Free

Want to use ChatGPT but want your AI to have Internet access? Google Gemini is the perfect solution for you.

So, what is Google Gemini, is there any difference in writing prompts for Google Gemini compared to ChatGPT? In this article, I, Lam Panda, will guide you in detail on how to use Google Gemini.

I updated this article from the article What is Google Bard, so if you search for What is Google Bard on Bing, you may be redirected to this article 🙂

Google Gemini and the technology platform

What is Google Gemini?

Google Gemini is an artificial intelligence chatbot developed by Google, based on the large language model (LLM) platform, specifically the Gemini model, and previously a combination of PaLM (Pathway Language Model) and LaMDA (Language Model for Dialogue Applications).

First launched in February 2023 as Bard, on February 1, 2024 (my birthday) they said goodbye to PaLM model and switched to Gemini model to support Bard. But then renamed to Gemini in May 2024 to match the platform or to make it less cluttered.

You can also find some other names of Gemini searched by users such as: Google Bard, Bard AI, Gemini AI, Google Gemini, Google Chatbot,…

If you’re a big AI follower, you’ll see that Google raced to release Gemini in less than 6 months to compete with OpenAI’s ChatGPT, so of course they’ll probably have a little more bugs.

Outstanding advantages of Google Gemini

• One of the most outstanding advantages of Gemini is the ability to access the internet. If you are used to ChatGPT, when switching to Gemini you will feel the air from the internet flowing into the AI’s answers 🙂
• Additionally, you can upload photos, files, and other types of documents to Gemini without having to use the paid version.
• You can also use the microphone to speak and write prompts hands-free on the web version, which is not available on ChatGPT.

• If you are a Chrome user, regularly use Google Docs, Gmail, and Google products in general, then Gemini is truly born to support these applications.
• When you ask Gemini to compose an email, it can Create a draft email in your Gmail , you compose other types of documents, you just need to click the Export to Documents button and it will run to your Google Docs,…
• What’s more special is that you can also check where Gemini’s answer content comes from, click the Verify answer button and Gemini will automatically search to check if the answer is correct or not.
• In short, Gemini has many advantages over other AI chatbots that you should try once (it’s addictive).
• Additionally, you can click the copy button directly from Gemini and paste it to Docs but still keep the same format as the answer without being converted to markdown like ChatGPT.

Google Gemini Capabilities

Google Gemini is designed to communicate and generate human-like text in response to a variety of questions and requests. The goal of Google Gemini is to serve as an innovative and useful collaborator that people can interact with using natural language.

Some of the key capabilities of Google Gemini include:

• Answer your questions about the world, like “What is the capital of France?” or “How does a car engine work?”
• Create creative text formats, such as poems, codes, scripts, music, emails, letters, etc.
• Translate languages, such as translating documents from English to Spanish.
• Write different types of creative content, such as blog posts or short stories.
• Help you with work or study, such as writing a research paper or solving a math problem.
• Or be a friend and companion, such as chatting with you about your day or helping you relax.
• Create texts, translate languages and write different types of creative content
• Assists users in various tasks, such as creating lists or generating ideas

And Google Gemini is suitable for a wide range of users such as writers, researchers, students and professionals, providing a powerful tool to boost productivity and creativity.

What are LaMDA, PaLM and Gemini?

If you want to understand more about how Gemini works, you can read the section on these big language models. In case you don’t need to get too deep, read on below to learn how Google Gemini works.

What is LaMDA?

LaMDA (Language Model for Dialogue Applications) is a large language model developed by Google Brain and specializes in dialogue applications.

LaMDA was initially introduced under the name Meena in 2020 and was later officially announced at the Google I/O 2021 event. LaMDA is built on the Transformer architecture, which allows the model to understand and generate long paragraphs of text, predicting the next word in a sentence based on context.

What is PaLM?

PaLM (Pathways Language Model) is another large language model from Google, with a training size of over 780 billion tokens and 540 billion parameters. First announced in April 2022 and accessible via API in March 2023.

Google also developed smaller versions of PaLM with 8 and 62 billion parameters to test the effect of model size. PaLM-E, an extension of PaLM, is a state-of-the-art vision-language model that can be used for robot control without retraining or fine-tuning.

In May 2023, Google announced PaLM 2 trained on 100 languages, 3.6 trillion tokens, and 340 billion parameters, a huge improvement over the previous version.

What is Gemini model?

Although PaLM is powerful, it still has certain limitations. Therefore, Google developed another model in parallel, Gemini , a multimodal model. This is a powerful model that can handle many different types of documents including images, audio, video, etc., which Google first announced in May 2023.

However, in May they were still developing, until February 1, 2024 they officially changed the Bard platform model from PaLM 2 to Gemini model. In May 2024 they changed the name of Bard to Gemini.

If you can’t imagine it, it’s similar to OpenAI’s GPT-4o model. Although Google was ahead, it was heavily criticized by the media and the shadow of ChatGPT was too big, so people just liked ChatGPT and criticized Gemini.

How does Google Gemini work?

Google Gemini works by leveraging natural language processing and machine learning to generate human-like text responses to user prompts. The interface of Google Gemini is also very simple and intuitive, making it easy for users to use when switching from ChatGPT or trying it for the first time.

Google Gemini’s Natural Language Processing Capabilities

Google Gemini’s Natural Language Processing (NLP) capabilities are powered by a combination of machine learning and NLP techniques, allowing it to generate high-quality, practical answers to user questions.

These capabilities help Google Gemini understand and generate human-like text, simulating a conversation with a user.

Some of Google Gemini’s key NLP capabilities include:

1. Natural Language Understanding (NLU) : Google Gemini can understand the meaning and context of user questions, thanks to being trained on large amounts of text data. This helps it generate relevant and informative answers.
2. Understanding Context : Google Gemini uses Google’s Language Model for Conversational Applications (LaMDA) to better understand user requests and apply context to its responses. This helps it maintain a consistent and contextual conversation with the user.
3. Text Generation : Google Gemini’s NLP capabilities allow it to generate a variety of content types, such as summarization, translation, and creative writing. This makes it a versatile tool for users looking for assistance with various writing tasks.
4. Language Translation : Google Gemini can translate text between multiple languages, thanks to being trained on multilingual data. This allows users to communicate with the AI in their preferred language and get translations for different texts.

Overall, Google Gemini’s NLP capabilities make it a powerful and versatile AI chatbot that can assist users in a variety of tasks, from information retrieval to creative content creation.

How to use Google Gemini

How to create an account and use Google Gemini

To understand how it works, we will need to interact with Google Gemini, just follow these steps:

Step 1: Go to Google Gemini and log in to your Google account.

Step 2: Now you just need to get familiar with the interface and you’re done.

To chat with Gemini, you just need to write a short content and ask for a specific task, and it will understand your intention and respond. If you write a prompt that is too detailed, Gemini will not understand or misunderstand 🙂 So keep it short.

Gemini’s interface is also very simple, the yellow frame on the left is where chats are stored and new chats are created.

Learn about add-ons in Google Gemini

Unlike ChatGPT, Gemini supports more things to help you focus on concise prompts and speed up the process of using AI including the places I marked in gold:

1. Write your command there
2. Can upload images to ask questions
3. If you are lazy, you can click on the microphone and speak instead of writing by hand.
4. The basic requests people use, it will appear after your first question.
5. You can create a share link for the chat, transfer to Docs , or create a draft email/
6. Clicking the Verify button will take you to Google Search to find content about and compare it with Gemini’s answer.
7. You can Copy the content of that chat.

Unlock Google Gemini’s Most Awesome Features

You may already know or soon will: you can ask Google Gemini to do more right in the Gemini interface like:

• Search, summarize Youtube videos
• Find hotels, flights
• Find address on Google Maps

Integrate directly into work activities such as:

• Search for emails. For example, ask for a summary of all emails received in the past week.
• Find documents in your Drive, Google Docs (this is not possible yet 🙂 but will be in the NEAR future).
• …

All you need to do is: Click on Install> Utilities> follow the instructions (just agree to continue) and finally you just need to press @ in the chat box to summon the above features.

For now you can use all the features, except for the Google Workspace part as it is still developing to dig up your documents.

How to use Google Gemini Advanced for free!

As promised at the beginning of the article, here I, Lam Panda, will guide you how to use Google Gemini Advanced for free, completely free, no need to buy anything!

Just log in to Google AI Studio then agree to their terms and conditions, and finally use it :3

If you are confused with the Google AI Studio homepage interface, you can go directly to the App section to write Prompt in AI Studio: https://aistudio.google.com/app/prompts/new_chat

Then you just need to choose Gemini 1.5 Flash or Pro model and use it comfortably.

Oh, within limits, okay 🙂 They don’t specify the quantity.

However, I tried chatting for nearly 2 hours with a bunch of questions and it didn’t say that it reached the limit. I guessed it would be around 500k tokens per day,… that’s probably it.

How to write prompt with Gemini

Information from official documents

Unlike ChatGPT, Claude, and other models, it seems Google designed Gemini to ask as short a question as possible and the AI will answer as fully as possible.

According to their official documentation, you only need to use 4 elements including: Role, Task, Context and Format.

It’s best to keep your question and role requirements concise, providing context elements preferably under 21 words, this is also according to their documentation.

Why should you do it? Because I tried it, I left it for a while to switch to Claude and ChatGPT, so when I came back, I had a huge framework asking for context roles,… And Gemini answered like hell 🙂

So I had to go back to the document they just published on how to use Gemini effectively and accurately to learn how to prompt again with this new Gemini model.

Document link: Gemini for Google Workspace Prompt Guide If you click here, you have to fill out a bunch of forms. It’s easier to view the PDF Prompt Guide and download it here.

3 golden rules in writing prompts for Gemini

After a little experimentation, I came up with the following factors to simplify the prompting problem so that it works effectively with Gemini. Basically, you just need to ensure the 3 factors in the “golden rule”: simple, request, and you know what the purpose of your question is, then everything will be fine.

If you still don’t know, here are the steps for you to write a prompt for Gemini, the order is based on the assessment of the importance of the factors according to the way I use and share with you:

1. Start simple : Start with a simple, clear prompt about your wants or requirements, for example: what is apple, which is better ChatGPT and Gemini 🙂
2. Use verbs in requests : Begin prompts with action verbs like “Write,” “Create,” or “Summary” instead of phrases like “Can you.”
3. Assign specific roles: assign roles so Gemini can identify the basic context of what you want such as: marketing specialist, Python programmer, project manager,…
4. Add context : Provide specific and relevant context about the task you want Gemini to perform, making it easier for it to understand your request.
5. Add output format requirements : Add other format requirements such as happy, sad, angry tone or add output format requirements such as: list, create table,…
6. Provide examples : Provide examples of the type of content you are looking for, this will help Gemini generate more relevant and accurate results, like ChatGPT but shorter.

Next, if you’ve done the 3 golden rules, you can add a specific role at the beginning of your prompt, so Gemini can understand more about your prompt.

Regarding rule number 4, you can add a brief specific context. But in practice, in my context section, it will be relatively complicated and when brought to Gemini, it will be a bit confusing and incomprehensible. So it is simpler to write it briefly in the role assignment section.

For example: a marketing specialist is planning a project.[ABC] for industry X customer[Gì đó] .

Just keep it short, marketing specialist is the role ; working on a project[ABC] for industry X customer[Gì đó] This is the context.

For more advanced level, you can read my article on the art of writing prompts to write prompts for Gemini. But remember: keep it short, don’t be like me and be long-winded or they won’t understand.

Also, you should chat with Gemini, point out the mistake and make him rewrite, don’t rush to create a new chat to make him reply from the beginning.

This is also another special point compared to ChatGPT, although ChatGPT has it, it is not clear about the development in the answer, after a short time it will reply with duplicate content or go away immediately.

More technically, Gemini has a wider memory context than the free version by tens of thousands of words, while the free ChatGPT only forgets about 1000 words 🙂

Examples of “gout” prompts and bad prompts

Good prompt examples for Google Gemini

1. Prompt: “What is Google Gemini blog outline?”
   • Good Reason: Short, focused “outline” format for the specific purpose of a “blog post” on a specific topic “What is Google Gemini”.
2. Prompt: “Write a brief summary of the main benefits of using solar energy.”
   • Good reasons: This prompt is clear, specific and focused on a single topic, has a requirement for the answer format, the content to be answered is also concise, Gemini will be less likely to create misinformation.
3. Prompt: “List five effective strategies to improve time management skills.”
   • Why it’s good: This prompt is simple and asks for a specific number of items, which helps guide the AI to generate a concise and organized response.
4. Prompt: “Explain the process of photosynthesis in simple language for middle school students.”
   • Why it’s good: This prompt provides context by specifying the target audience, allowing Google Gemini to tailor the response to the appropriate level of complexity.
5. Prompt: “Compare and contrast the advantages and disadvantages of electric cars versus gasoline cars.”
   • Good reason: This prompt is clear and specific, asking for a comparison between two separate topics, which helps Google Gemini generate a structured and informative answer.

Example of bad prompt Google Gemini

Note: this is not a good example, it is not 100% bad , it means it can be improved a little to make it better:

1. Prompt: “Help me outline my blog post about What is Google Gemini?
   • Reasons why it’s not okay: it’s not necessary to chat, you can ask directly, you can shorten the sentence to avoid Gemini misunderstanding your meaning.
2. Prompt: “Tell me something.”
   • Why it’s bad: This prompt is vague and lacks context, making it difficult for the AI to generate relevant and informative responses.
3. Prompt: “What’s the best thing ever?”
   • Why it’s not good: This prompt is subjective and open-ended, making it difficult for the AI to provide specific and meaningful answers.
4. Prompt: “Write a story about a cat and a dog.”
   • Why it’s bad: This prompt is too broad and lacks specific details or context, making it difficult for the AI to create a cohesive and engaging story.
5. Prompt: “Can you help me with my homework?”
   • Why it’s not working: This prompt is unclear and doesn’t provide information about the specific exercise task, making it difficult for the AI to provide relevant assistance.
6. Prompt: “What should I do today?”
   • Why it’s not good: This prompt is vague and lacks context about the user’s interests, priorities or schedule,… If you don’t know what to do, how will AI know 🙂 Hmm, you can also think more openly and go find ideas (I’m not saying I ask it like that all the time).

The next thing you need to do is log in to Gemini and try it out, just log in with your Google account and use Gemini AI. After logging in, try the “golden rule” to upgrade your prompt skills.

Advantages and disadvantages of Google Gemini

Advantages of Google Gemini

1. Internet Accessibility : Google Gemini provides the latest and updated results by extracting information from the web through Google search, making it beneficial for research and information gathering purposes.
2. Natural Language Understanding : Google Gemini can understand commands entered through text or voice in natural language and create a human-like conversation, making it similar to virtual assistants like Google Assistant or Siri.
3. Exceptional Content Creation Capabilities : For users of Google products, Gemini is almost perfectly compatible to serve their work.
4. Huge memory capacity: GPT-4o has only 128k contexts, Claude 3 has 200k contexts and Gemini Advanced can remember up to 1 million tokens! That’s nearly 1500 a4 pages of text!

Limitations of Google Gemini

1. Biased, Inaccurate and Ambiguous Information : One of the biggest drawbacks of Google Gemini is its scope for providing biased, inaccurate and ambiguous information.
2. Still in development : Google Gemini is still in development, which means it may have some performance limitations and may not be able to handle all tasks perfectly. Specifically and intuitively, it is the name of the person who emailed me that needs no further proof: Gemini, an early AI experiment by Google … Ok… They still seem unsure about their product but have to release it.
3. Creative Limitations : While Gemini is capable of creative writing, it cannot produce long articles, essays, or detailed works of fiction such as a complete story with proper phrasing. Of course, because it is very robotic, as all AIs are.
4. Repetitive and unoriginal content : While Gemini can produce useful content, the content can be repetitive and unoriginal. My own testing shows that difficult questions and questions that require too much listing will cause the AI chatbot to generally produce incorrect answers and repeat answers.

In short, it’s still bad at the things that other AIs are bad at, it’s just that the media exaggerates it and makes it sound like Gemini is bad. Actually, it’s not that bad, it’s actually quite good.

If you want a more direct look, I will compare them next below.

Google Gemini vs ChatGPT vs Microsoft Copilot

Google Gemini, ChatGPT, and Microsoft Copilot are all AI chatbots designed to assist users with different tasks. Each chatbot has its own pros and cons, making them suitable for different use cases.

To find out which tool is best for you or your job right now, let’s take a look at all 3 AI chatbots that are leading the market.

Some Key Points About Google Gemini, ChatGPT, and Microsoft Copilot

Google Gemini

• Based on the latest Gemini 1.0 model
• Can upload images and receive commands as voice
• Good at finding accurate and up-to-date information in real time
• Provides creative content creation and language translation capabilities
• paid will be able to use the much better Gemini 1.5 model and context increased to 1 million tokens and many other benefits such as: Unlimited Google Meet meetings, 2TB Drive storage, direct integration into your Workspace,…

ChatGPT

• Using OpenAI’s GPT-3.5 Turbo and GPT-4o models
• Text-based AI chatbot generates coherent and contextually relevant responses
• Good at data analysis and problem solving
• A ChatGPT Plus subscription is required to access more models, which costs $20 per month. You can create additional images, create your own bot, and more.

Microsoft Copy

• Integrated into Microsoft’s Bing search engine and uses GPT-4 technology
• Less complex than ChatGPT but can still answer questions and perform simple tasks
• Can generate images and retrieve accurate information
• Free to use limited version, when upgraded for $20/month you will receive unlimited access benefits, increase the number of photos to 100/day,..

Comparison table

This comparison table is based on my personal opinion and my evaluation through using all of them for nearly 2 years now, so it may be different from your usage and evaluation.

Features	Google Gemini	ChatGPT	Microsoft Copy
Total visits (1 month)*	Over 189 million	Over 1.5 billion	Over 1.2 billion**
Release date	March 21, 2023	November 30, 2022	February 22, 2023
Usage model	Gemini 1.0	GPT-3.5 turbo	GPT-4
Main uses	Provide general information about the problem	Data analysis, problem solving	Answer questions and search
Cost of use	Free and Advanced 20$/month	Free and Plus 20$/month	Free and Pro 20$/month
Suitable for	The task requires factual and up-to-date information.	Creative writing, inductive reasoning	Retrieve information and images on the web

*Data from Similarweb

**In which, I am not sure about the number because there are both normal searchers and other searchers 🙂 and Bing Chat, Bing Image Creator are also on Bing’s homepage and not separated so this number is Bing’s total traffic…

Intended use

The intended use of each AI tool varies based on their power and capabilities:

• Google Gemini is designed to answer logical, mathematical, and programming questions accurately. It is good at finding accurate and real-time information.
• ChatGPT generates coherent, contextually relevant answers that simplify the problem. It is suitable for creative writing, reasoning, coding, and data analysis.
• Microsoft Copilot aims to provide information by leveraging its vast knowledge base of search engines. It is best for pulling information from the web and creating images.

Each tool has its own unique strengths, and the best one for you depends on your specific use case. Ultimately, you should still try and see which one works for you.

Here’s my summary of how to use them effectively:

• Google Gemini : suitable for work-related, creative and less complex tasks that require cross-referencing with online documents.
• ChatGPT is suitable for most people, has the ability to reason and answer complex questions, and is suitable for both work and entertainment.
• Microsoft Copilot : it is best to get information from web and create free images with high quality.

Wrap up the dress

In short, Google Gemini is a versatile and powerful AI tool that can assist users in a variety of tasks. It will continue to grow and improve, and its potential applications and capabilities are expected to expand, making it even more powerful in the future.

Anyway, Google has plenty of data but lacks time to develop… Well, I don’t know. But it will definitely be better, instead of just using ChatGPT, you can combine it with Gemini to cross-check the answers and find the best answer for yourself.

If there is any wrong content in my article or you want to add more comments, please comment below so I can absorb it immediately, thank you for reading such a long article.

FAQ about Google Gemini