How to get your Gemini API key

 

How to get your Gemini API key

Connecting Gemini—Google’s LLM and formerly known as Bard—with your internal applications and/or product can fundamentally change how your employees interact with your applications and/or how customers use your product. 

But before you can access one of Gemini’s models via API requests, you’ll need to generate an API key within the Google AI Studio. 

We’ll help you do just that by walking through each of the 5 steps you’ll need to take.

Step-by-Step Guide to Generating Your Gemini API Key

1. Sign in to Google AI Studio

Navigate to aistudio.google.com. You must sign in using a standard Google Account (Gmail).

Note: If you are using a Google Workspace account (for work or school), your administrator may need to enable access to "Google AI Studio" in the Admin Console.

2. Navigate to the API Key Section

Once you are logged into the dashboard, look at the left-hand sidebar menu. Click on the button labeled "Get API key" (often identified by a key icon 🔑)

3. Create the Key

You will see two primary options for creating a key:

• Create API key in new project: This is the easiest method. Google will automatically create a new Google Cloud project for you in the background and generate a key for it.

• Create API key in existing project: Choose this if you already have a Google Cloud project (e.g., for Firebase or other Google APIs) and want to keep your resources organized together.

Click "Create API key" after making your selection.

4. Copy and Secure Your Key

A pop-up will appear containing your unique API key.

1. Copy the key immediately.

2. Store it safely: Use a password manager or an environment variable file (.env).

3. Security Warning: Never hard-code this key directly into public GitHub repositories. If your key is leaked, anyone can use your quota, which may lead to service interruptions or charges if you have billing enabled.

---

Managing and Restricting Your Key

To ensure your key is used correctly, you can manage it further:

Action	How to do it
Delete/Rotate	Click the trash icon next to the key in the "API keys" list and generate a new one.
Set Billing	Go to the "Usage & Billing" tab to upgrade to the Pay-as-you-go tier for higher rate limits.
Restrictions	Click on the Project ID link to open the Google Cloud Console, where you can restrict the key to only work with specific IP addresses or websites.

---

Quick Connection Test

Once you have your key, you can verify it works by running a simple curl command in your terminal (replace YOUR_API_KEY with your actual key):

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent?key=YOUR_API_KEY" \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
      "contents": [{
        "parts":[{"text": "Write a tagline for a coffee shop."}]
      }]
    }'

Part 2: Detailed Guide (Manual Project & Key Setup)

Use this if you want to link your key to a specific project (e.g., "Trend-Analysis-2026") 
for better organization in the Google Cloud Console.

Step 1: Create a Project in Google Cloud

Go to the Google Cloud Console.
Click the Project Dropdown (top left) > New Project.
Enter your Project Name and click Create.

Step 2: Enable the Gemini API

In the Cloud Console search bar, type "Generative Language API".
Select it from the results and click Enable.

Step 3: Link Project to AI Studio

Return to Google AI Studio.
Click "Get API key" in the sidebar.
Click "Create API key in existing project".
A list will appear; select the project you just created in Step 1.
Click "Generate Key".

Part 3: Securing and Testing Your Key

1. Copy and Store

Once the key is generated, copy it immediately.

• Recommendation: Store it in a .env file for your Python scripts:

  Bash

  # .env file
  GEMINI_API_KEY=your_actual_key_here
  

2. Verify with Gemini 3 (2026 Update)

The latest SDKs (like google-genai) use the key automatically if stored in your environment. You can test it with this minimal Python snippet:

Python

from google import genai

# The client automatically looks for the GEMINI_API_KEY environment variable
client = genai.Client()

response = client.models.generate_content(
    model="gemini-3-flash",
    contents="Hello, confirm API key is active."
)

print(response.text)

---

Critical 2026 Security Reminders

• Do Not Hardcode: Never put your key inside client = genai.Client(api_key="...") if you plan to share your code or push it to GitHub.

• Usage Limits: Free tier keys now include SynthID watermarking on all generated images and media by default.

• API Key Restrictions: If you want to prevent others from using your key, go to Google Cloud Console > APIs & Services > Credentials, click your key, and restrict it to "Generative Language API" only.

 

Scientists Develop Universal Brain-Computer Interface That Lets People Play Games With Just Their Thoughts

Advancing Accessibility Through Self-Calibrating BCIs

Typically, these devices require extensive calibration for each user — every brain is different, both for healthy and disabled users — and that has been a major hurdle to mainstream adoption. This new solution can quickly understand the needs of an individual subject and self-calibrate through repetition. That means multiple patients could use the device without needing to tune it to the individual.

“When we think about this in a clinical setting, this technology will make it so we won’t need a specialized team to do this calibration process, which is long and tedious,” 

“It will be much faster to move from patient to patient.”

Implementation and Results

The subjects wear a cap packed with electrodes that is hooked up to a computer. The electrodes gather data by measuring electrical signals from the brain, and the decoder interprets that information and translates it into game action.

Millán’s work on brain-computer interfaces helps users guide and strengthen their neural plasticity, the ability of the brain to change, grow, and reorganize over time. These experiments are designed to improve brain function for patients and use the devices controlled by brain-computer interfaces to make their lives easier.

In this case, the actions were twofold: the car racing game, and a simpler task of balancing the left and right sides of a digital bar. An expert was trained to develop a “decoder” for the simpler bar task that makes it possible for the interface to translate brain waves into commands. The decoder serves as a base for the other users and is the key to avoiding the long calibration process.

The decoder worked well enough that subjects trained simultaneously for the bar game and the more complicated car racing game, which required thinking several steps ahead to make turns.

The researchers called this work foundational, in that it sets the stage for further brain-computer interface innovation. This project used 18 subjects with no motor impairments. Eventually, as they continue down this road, they will test this on people with motor impairments to apply it to larger groups in clinical settings.

“On the one hand, we want to translate the BCI to the clinical realm to help people with disabilities; on the other, we need to improve our technology to make it easier to use so that the impact on these people with disabilities is stronger.

On the side of translating the research, Millán and his team continue to work on a wheelchair that users can drive with the brain-computer interface. At the South by Southwest Conference and Festivals this month, the researchers showed off another potential use of the technology, controlling two rehabilitation robots for the hand and arm. This was not part of the new paper but a sign of where this technology could go in the future. Several people volunteered and succeeded in operating the brain-controlled robots within minutes.

“The point of this technology is to help people, help them in their everyday lives,” 

 “We’ll continue down this path wherever it takes us in the pursuit of helping people.”

What is 6G? Overview of 6G networks & technology

 

What is 6G?

6G (sixth-generation wireless) is the successor to 5G cellular technology. 6G networks will be able to use higher frequencies than 5G networks and provide substantially higher capacity and much lower latency. One of the goals of the 6G internet is to support one microsecond latency communications. This is 1,000 times faster -- or 1/1000th the latency -- than one millisecond throughput.

The 6G technology market is expected to facilitate large improvements in the areas of imaging, presence technology and location awareness. Working in conjunction with artificial intelligence (AI), the 6G computational infrastructure will be able to identify the best place for computing to occur; this includes decisions about data storage, processing and sharing.

It is important to note that 6G is not yet a functioning technology. While some vendors are investing in the next-generation wireless standard, industry specifications for 6G-enabled network products remain years away.

What are the advantages of 6G vs. 5G?

6G networks will operate by using signals at the higher end of the radio spectrum. It is too early to approximate 6G data rates, but Dr. Mahyar Shirvanimoghaddam, senior lecturer at the University of Sydney, suggested a theoretical peak data rate of 1 terabyte per second for wireless data may be possible. That estimate applies to data transmitted in short bursts across limited distances. LG, a South Korean company, unveiled this type of technology based on adaptive beamforming in 2021.

This level of capacity and latency will extend the performance of 5G applications. It will also expand the scope of capabilities to support new and innovative applications in wireless connectivity, cognition, sensing and imaging. With 6G, access points will be able to serve multiple clients simultaneously via orthogonal frequency-division multiple access.

6G's higher frequencies will enable much faster sampling rates than with 5G. They will also provide significantly better throughput and higher data rates. The use of sub-mm waves -- wavelengths less than 1 millimeter -- and frequency selectivity to determine relative electromagnetic absorption rates is expected to advance the development of wireless sensing technology.

Mobile edge computing will be built into all 6G networks, whereas it must be added to existing 5G networks. Edge and core computing will be more integrated as part of a combined communications and computation infrastructure framework by the time 6G networks are deployed. This approach will provide many potential advantages as 6G technology becomes operational. These benefits include improved access to AI capabilities and support for sophisticated mobile devices and systems.

When will 6G internet be available?

6G internet is expected to launch commercially in 2030. The technology makes greater use of the distributed radio access network (RAN) and the terahertz (THz) spectrum to increase capacity, lower latency and improve spectrum sharing.

While some early discussions have taken place to define the technology, 6G research and development (R&D) activities started in earnest in 2020. 6G will require development of advanced mobile communications technologies, such as cognitive and highly secure data networks. It will also require the expansion of spectral bandwidth that is orders of magnitude faster than 5G.

China has launched a 6G test satellite equipped with a terahertz system. Technology giants Huawei Technologies and China Global reportedly plan similar 6G satellite launches in 2021. Many of the problems associated with deploying millimeter wave radio for 5G must be resolved in time for network designers to address the challenges of 6G.

How will 6G work?

It's expected that 6G wireless sensing solutions will selectively use different frequencies to measure absorption and adjust frequencies accordingly. This method is possible because atoms and molecules emit and absorb electromagnetic radiation at characteristic frequencies, and the emission and absorption frequencies are the same for any given substance.

6G will have big implications for many government and industry approaches to public safety and critical asset protection, such as the following:

• threat detection;
• health monitoring;
• feature and facial recognition;
• decision-making in areas like law enforcement and social credit systems;
• air quality measurements;
• gas and toxicity sensing; and
• sensory interfaces that feel like real life.

Improvements in these areas will also benefit smartphone and other mobile network technology, as well as emerging technologies such as smart cities, autonomous vehicles, virtual reality and augmented reality.

Do we even need 6G?

There are a number of reasons we need 6G technology. They include the following:

• Technology convergence. The sixth generation of cellular networks will integrate previously disparate technologies, such as deep learning and big data analytics. The introduction of 5G has paved the way for much of this convergence.
• Edge computing. The need to deploy edge computing to ensure overall throughput and low latency for ultrareliable, low-latency communications solutions is an important driver of 6G.
• Internet of things (IoT). Another driving force is the need to support machine-to-machine communication in IoT.
• High-performance computing (HPC). A strong relationship has been identified between 6G and HPC. While edge computing resources will handle some of the IoT and mobile technology data, much of it will require more centralized HPC resources to do the processing.

Who is working on 6G technology?

The race to 6G is drawing the attention of many industry players. Test and measurement vendor Keysight Technologies has committed to its development. Major infrastructure companies, such as Huawei, Nokia and Samsung, have signaled that they have 6G R&D in the works.

The race to reach 5G may end up looking minor when compared with the competition to see which companies and countries dominate the 6G market and its related applications and services.

The major projects underway include the following:

• The University of Oulu in Finland has launched the 6Genesis research project to develop a 6G vision for 2030. The university has also signed a collaboration agreement with Japan's Beyond 5G Promotion Consortium to coordinate the work of the Finnish 6G Flagship research on 6G technologies.
• South Korea's Electronics and Telecommunications Research Institute is conducting research on the terahertz frequency band for 6G. It envisions data speeds 100 times faster than 4G Long-Term Evolution (LTE) networks and five times faster than 5G networks.
• China's Ministry of Industry and Information Technology is investing in and monitoring 6G R&D in the country.
• The U.S. Federal Communications Commission (FCC) in 2020 opened up 6G frequency for spectrum testing for frequencies over 95 gigahertz (GHz) to 3 THz.
• Hexa-X is a European consortium of academic and industry leaders working to advance 6G standards research. Finnish communications company Nokia is leading that project, which also includes Ericsson, a Swedish operator, and TIM in Italy.
• Osaka University in Japan and Australia's Adelaide University researchers have developed a silicon-based microchip with a special multiplex to divide data and enable more efficient management of terahertz waves. During testing, researchers claimed the device transmitted data at 11 gigabits per second compared to 5G's theoretical limit of 10 Gbps of 5G.

Future scope of 6G networks

About 10 years ago, the phrase "Beyond 4G" (B4G) was coined to refer to the need to advance the evolution of 4G beyond the LTE standard. It was not clear what 5G might entail, and only pre-standards R&D-level prototypes were in the works at the time. The term B4G lasted for a while. It referred to what could be possible beyond 4G. Ironically, the LTE standard is still evolving, and 5G will use some aspects of it.

Similar to B4G, Beyond 5G is seen as a path to 6G technologies that will replace fifth-generation capabilities and applications. 5G's many private wireless communications implementations involving LTE, 5G and edge computing for enterprise and industrial customers have helped lay the groundwork for 6G.

Next-generation 6G wireless networks will take this one step further. They will create a web of communications providers -- many of them self-providers -- much in the way that photovoltaic solar power has brought about cogeneration within the Smart grid. 6G could advance mesh networks from concept to deployment, helping to extend coverage beyond the range of older cell towers.

Data centers are already faced with big 5G-driven changes. These include virtualization, programmable networks, edge computing and issues surrounding simultaneous support of public and private networks. For example, some business customers may want to combine on-premises RAN with hybrid on-premises and hosted computing -- for edge and core computing, respectively -- and data center-hosted core network elements for private business networks or alternative service providers.

6G radio networks will provide the communication and data gathering necessary to accumulate information. A systems approach is required for the 6G technology market that makes use of data analytics, AI and next-generation computation capabilities using HPC and quantum computing.

In addition to profound changes within RAN technology, 6G will bring changes to the core communications network fabric as many new technologies converge. Notably, AI will take center stage with 6G.

Other changes 6G is likely to bring include the following:

• Nano-core. A so-called nano-core is expected to emerge as a common computing core that encompasses elements of HPC and AI. The nano-core does not need to be a physical network element. Instead, it could encompass a logical collection of computational resources, shared by many networks and systems.
• Edge and core coordination. 6G networks will create substantially more data than 5G networks, and computing will evolve to include coordination between edge and core platforms. In response to those changes, data centers will have to evolve.
• Data management. 6G capabilities in sensing, imaging and location determination will generate vast amounts of data that must be managed on behalf of the network owners, service providers and data owners.

What is a 7G network and why is it needed?

Even though 6G networks are not expected to be operational until at least 2032, research has started on seventh-generation (7G) wireless technologies. The IEEE, through its Extremely High Throughput working group, is developing the 802.11be specification for 7G and an industry certification in conjunction with the Wi-Fi Alliance.

The IEEE's amended standard is expected in May 2024. It will provide device manufacturers with design specifications to govern interoperability and performance.

6G networks are attempting to extend fast Gigabit Ethernet connectivity to commercial and consumer devices. 6G is expected to provide substantially higher throughput and data flow. As envisioned, 6G will enable the following:

• deliver a theoretical data rate of about 11 Gbps simultaneously across multiple gigahertz channels;
• deploy up to three 160-megahertz (MHz) bandwidth channels; and
• multiplex up to eight spatial streams.6GE -- the "E" stands for extension -- is an interim step between 6G and 7G that will use a newly licensed 6 GHz channel that extends the available frequencies used to transmit 6G signals. The FCC in 2020 was the first regulatory body to greenlight the 6 GHz spectrum to help foster innovation of 6GE Wi-Fi devices.

• 7G technology will represent a quantum leap in bandwidth to support ultradense workloads. For example, 7G has the potential to enable continuous global wireless connectivity via integration in satellite networks for earth imaging, telecom and navigation. Enterprises could implement 7G to automate manufacturing processes and support applications that require high availability, predictable latency or guaranteed quality of service.

  Compared to 6G, 7G is designed to do the following:

  • deliver data up to 46 Gbps -- more than four times the rate of 6G projection;
  • double the size of the channel to 320 MHz; and
  • afford 16 spatial streams, compared to eight in 6G.