Google Gemini explained: Everything you need to know about Google’s next-g en AI
When you buy through links on our articles, 未来 and its syndication partners may earn a （売買）手数料，委託（する）／委員会／権限.
Along with OpenAI's ChatGPT, Microsoft's Copilot and Apple 知能, Google Gemini is one of the 支配的な 軍隊s in the world of 人工的な 知能 (AI) and chatbots. Gemini is both the 指名する for Google chatbot and the LLM that 力／強力にするs it, and it's 解放する／自由な to use 経由で a web browser, or on your 動きやすい, but there's a paid-for 見解／翻訳／版 called Gemini 前進するd that has some juicy features, like AI ビデオ 世代 with Veo 2.
Google Gemini burst の上に the scene in February 2024 and すぐに made some big waves in the AI world, but it was the 解放（する） of Gemini Live at the Made for Google event in August 2024 that really put it on the 地図／計画する. Gemini Live is a 見解／翻訳／版 of Gemini that runs on 動きやすい phones, even the iPhone, and enables you to have 解放する／自由な flowing conversations about コンビナート／複合体 topics using your 発言する／表明する instead of having to type on the keyboard.
But is Gemini Live enough to 敗北・負かす Apple's AI-高めるd Siri (if and when it finally arrives) or ChatGPT's 前進するd 発言する／表明する 方式? What can it do 権利 now, and what about in the 未来? And if you want to use Gemini, how 正確に／まさに do you do that? We’ve taken a 深い dive into the world of Gemini to find the answers to all these questions and more. If you’re curious about Google’s 最新の AI 成果／努力s, this is the place to be.
What is Google Gemini?
Gemini is Google’s large language model (LLM), which is 現在／一般に up to 見解／翻訳／版 2.5 プロの／賛成の. What’s an LLM? It’s the system that underpins the types of AI 道具s you’ve probably seen and interacted with on the internet. For example, GPT-4 力／強力にするs ChatGPT-4o, OpenAI’s chatbot.
Gemini is more than an AI model, though, as it’s also the 指名する and 身元 for it's chatbot, 以前 known as 妨げる／法廷，弁護士業d. Essentially, Google has 簡単にするd things by calling both the underlying model and chatbot itself Gemini.
その上に, there’s a 解放する／自由な Gemini app for both Android and iOS now, and Gemini can 取って代わる Google Assistant on your Android phone if you wish. There's also a 解放する／自由な 見解／翻訳／版 of Google Gemini that is accessible 経由で any internet browser.
On 最高の，を越す of all this, Google has rebranded its Duet AI service, 目的（とする）d at 商売／仕事s, as Gemini for Workspace, with a whole bunch of 生産性-関係のある chops on 申し込む／申し出.
The final 新たな展開 is that 同様に as the basic (解放する／自由な) 見解／翻訳／版 of Gemini for 消費者s, there is also 賞与金 申し込む／申し出ing for the AI known as Gemini 前進するd, which is 利用できる through a subscription called Google One AI 賞与金. Gemini 前進するd gives you more 接近 to Gemini's more 有能な LLMs, and more 接近 to the Gemini 深い 研究 feature.
To sum up, all Google’s AI 所有物／資産／財産s are now under the Gemini umbrella to 簡単にする things, whether that’s AI for 消費者s or 商売／仕事s, and whether 接近ing Gemini 経由で the web, or the assistant or app on your smartphone.
What can Gemini do?
The short answer to this question is: a lot. But you probably want us to (a)手の込んだ／(v)詳述する a bit more than that.
As we’ve just discussed, Gemini is an expansive umbrella for a whole lot of AI features and functionality 配達するd 経由で different avenues. It has the same generative 能力s as other chatbots, like ChatGPT, so if you tell Gemini where you are going on your next trip, it will be able to help you pack. Or ask it to explain who Socrates was and sit 支援する for a history lesson. It can answer questions, 要約する text, 令状 code and translate.
Gemini is a multimodal LLM. In other words, it can を取り引きする さまざまな forms of input and 生産（高）, 含むing text, code, 音声部の, images and ビデオs. That gives it a lot of 柔軟性 to 成し遂げる a wide 範囲 of 仕事s.
For instance, you can upload an image (or take one with your smartphone) and Gemini can 分析する the image and tell you things about it. Paste some code into the Gemini 誘発する box and ask it to rewrite it and Gemini can do that.
Recently the ability to edit images has also been 追加するd to Gemini, so you can upload an image, and ask it to change it in some way.
There are four separate LLMs that Google has 器具／実施するd with its AI. They are Gemini Ultra, プロの／賛成の, Flash and Nano. If you go to the Gemini website and ask the chatbot a question you'll be using Gemini 2.0 or 2.5 Flash, which is Google's lightweight model, optimized for 速度(を上げる） and efficiency.
If you are a member of the paid tier 経由で Gemini 前進するd then you get 接近 to Gemini 2.5 プロの／賛成の, which is the largest model, and good for 高度に コンビナート／複合体 仕事s.
Finally, there's the 深い 研究 with 2.5 プロの／賛成の model, which is a 研究 スパイ／執行官, 有能な of 成し遂げるing in-depth 研究 on a 支配する for you, while you're 解放する／自由な to go off and do something else.
You can also talk to Gemini with your 発言する／表明する and have it talk 支援する by selecting the "Go Live" 選択 when chatting to the LLM. This 作品 best on the new Pixel 9 phones, which have the ability to 株 their 審査する and camera with Gemini, so it can 'see' what you're looking at and you can ask it questions about it.
Google has developing its own answer to Midjourney in Imagen 3, which is 統合するd into Gemini, too. You can use Imagen 3 through Gemini, by asking it to 簡単に create an image of a scene or person. Google is also developing native image 世代 for Gemini, which we'd 推定する／予想する to be 追加するd very soon.
Recently AI ビデオ 世代 has also been 追加するd using Veo 2. For this to work you need to be a Gemini 前進するd 加入者.
If you're a Gemini 前進するd 加入者 then you'll also find that Gemini is 麻薬中毒の into the likes of Gmail, Google 地図／計画するs, and YouTube. So, if you ask it for sightseeing 推薦s, for example, it will 旗 them up in Google 地図／計画するs.
These are some nifty abilities for sure, and those with Android smartphones get even more mileage for 解放する／自由な with the Gemini app. As already について言及するd, this can 取って代わる Google Assistant on an Android 装置, if you wish, running queries and working its AI 魔法 経由で other Google services (地図／計画するs, search, even Assistant itself, which isn’t 現実に chucked off your phone as such, and still lurks underneath).
That choice may lead you to wonder: is Gemini a good 交替／補充 for Google Assistant? 井戸／弁護士席, we covered this extensively in our 手渡すs-on with Gemini on Android, and there are certainly drawbacks with the AI. Gemini is noticeably slower than Google Assistant, and the AI still has bugs.
That said, the 初期の Gemini on Android experience was a lot wonkier than it is now. An example that springs to mind is that Gemini was supposed to を取り引きする smart home gadgets, but failed in that 仕事 when we tried it 最初 - but more recently, this feature has come good and now 作品.
It seems that Google is アイロンをかけるing out the problems with Gemini on 動きやすい pretty 速く, which is heartening to see, and the results with the Gemini app can be impressive. There are still wrinkles to smooth over ーに関して／ーの点でs of 取って代わるing Google Assistant on Android, but that should come in time.
全体にわたる, then, the freebie 見解／翻訳／版 does give you a lot to get on with, 特に for Pixel 9 使用者s. However, the paid take on Gemini is 現実に far more in-depth.
For those who use Google’s 生産性 apps and have a Google Workspace account, Gemini is 利用できる in Google Docs, Gmail, and more.
When was Gemini 解放（する）d?
Google Gemini was 解放（する）d on February 8 2024, with Google 確認するing that it 取って代わるd 妨げる／法廷，弁護士業d which was put out to pasture. Gemini was 利用できる すぐに, in both its 解放する／自由な form, and the paid 見解／翻訳／版, Gemini 前進するd. Google also began the rollout of the Android app in the US straight away.
The Gemini app is also 利用できる for Samsung phones, read our Samsung 星雲 Z Flip 6 review, and for iPhone and iPad.
Is Google Gemini 解放する／自由な?
The 基準 見解／翻訳／版 of Google Gemini is 解放する／自由な, but it’s more 限られた／立憲的な than the paid spin on the AI. Those who 支払う／賃金 a subscription for Gemini 前進するd get a lot more depth ーに関して／ーの点でs of 接近 to Gemini's more 前進するd 能力s 同様に as AI ビデオ 創造.
How much is Gemini 前進するd? Google 告発（する），告訴（する）／料金s $19.99 (£18.99 / AU$32.99) per month, but you can try it out for a 限られた／立憲的な time for 解放する／自由な, as a one-month 裁判，公判 is 現在／一般に 利用できる. The subscription comes with other 利益s, though, as Gemini 前進するd is 現実に part of the Google One AI 賞与金 計画（する）, which also 供給するs 2TB of cloud 貯蔵 の中で other extras.
Given that Google One with 2TB of 貯蔵 already costs $9.99 / £7.99 / AU$12.49 per month, if you need a cloud 貯蔵 locker, Gemini 前進するd looks a more tempting value proposition.
How do I use Google Gemini?
The way you use Google Gemini depends on 装置 you're using.
On a PC or laptop, you can 長，率いる to the Gemini website and use the AI there for 解放する／自由な in the typical way you interact with an online chatbot. If you subscribe to Gemini 前進するd, you get the 十分な AI experience 含むing AI ビデオ 世代, 詐欺師 返答s, more 接近 to 前進するd LLMs for 取引，協定ing with コンビナート／複合体 仕事s and 需要・要求するing creative needs, and the other 利益s we’ve 輪郭（を描く）d above.
On 動きやすい you can use the Gemini app on your Android phone (or 取って代わる Google Assistant with Gemini, as 公式文書，認めるd) or on iOS where you get all the same features you'll find in the browser, but with the 新規加入 of Gemini Live for conversing to Gemini using your 発言する／表明する.
If you are on a smartphone like the Google Pixel 9, then you can use Gemini Live with 高めるd 審査する 株ing and camera 支配（する）／統制する features.
Gemini vs GPT-4: what’s the difference?
How does Gemini stack up against GPT-4 in the 戦う／戦い of the large language models?
For one thing, when Gemini was first 明らかにする／漏らすd, Google （人命などを）奪う，主張するd it’s more 前進するd than GPT-4. In a blog 地位，任命する, Google showed results from eight text-based （判断の）基準s, with Gemini winning in seven of those 実験（する）s. Across 10 multimodal （判断の）基準s, Gemini （機の）カム out on 最高の，を越す in every one, によれば Google at least.
That would se em to 暗示する that Gemini is the superior system, but it’s not やめる so straightforward, of course. GPT-4 （機の）カム out in March 2023, and Gemini 1.5 プロの／賛成の （機の）カム out in February and Gemini 1.5 Flash （機の）カム out in May 2024. ChatGPT now has many more 前進するd models like 4o and o3, not to について言及する the new ChatGPT-4.5. Gemini is now up to 2.5 Flash and 2.5 プロの／賛成の models.
On the whole, features are 概略で 類似の between ChatGPT 加える and Gemini 前進するd, and both have a 解放する／自由な tier. It may 簡単に come 負かす／撃墜する to personal choice and finding the chatbot that fits your person style best.