5 Gemini features ChatGPT still can’t match
When you buy through links on our articles, Future and its syndication partners may earn a commission.
Credit: Shutterstock
Over the past few years we've seen ChatGPT and Gemini get some really useful features. Each AI assistant offers unique abilities and both ChatGPT and Gemini can handle many of the same jobs. But their differences become clearer when you move past the question and answer features.
Google has connected Gemini to services and devices it already controls, including Android and Google Photos. It has given Gemini a few creative tools that ChatGPT still doesn't offer, too. You can create a finished song, turn a research report into a podcast and build a reusable AI version of yourself.
I use both assistants regularly, and these are five things Gemini can currently do that ChatGPT cannot do natively. Of course, availability of these Gemini features depends on your device, account, subscription and location, so you may not see every feature yet.
1. Create an original song you can download
ChatGPT can write lyrics, suggest chord progressions and help plan a song. But it still does not have the ability to take that prompt and hand you a finished track with music and vocals.
Gemini can because it offers Lyria 3 music generator directly into the app. You can describe the song you want or upload a photo or video to use as inspiration. You can even upload text messages and ask it to create a song from those. The possibilities are endless right in the chatbot as Gemini then generates the music and lyrics together.
A prompt could be as simple as: "Create an upbeat pop song about surviving the first week of school as a tired parent."
You can download the finished result as an MP3 or save it as an MP4 with cover art. According to Google's instructions , generating a full track requires switching Gemini to its Pro model. This is one of the clearest gaps between the two assistants. ChatGPT can help you plan the song. Gemini can make something you can actually play.
2. Search through your Google Photos without uploading anything
Credit: Google
Finding one particular picture in a library containing thousands of nearly identical family photos can take forever. But Gemini can search a connected Google Photos account using the people, places and objects that appear in your images. You could ask it to find photos from a particular birthday, show you pictures from last summer or look for images containing a specific person.
It can answer questions using details from those photos, too. Google suggests prompts asking Gemini to identify the themes used at someone's previous birthday parties or recall what you ate during a vacation.
ChatGPT can analyze any photo you upload, but it cannot natively search through your entire Google Photos library. You have to find the relevant pictures and add them to the conversation yourself.
There are several restrictions. Google Photos access in Gemini requires a personal Google account, isn't available to users under 18 and only works in supported regions. You must turn on Gemini's Keep Activity setting and certain Google Photos features, including Face Groups.
I'd think carefully before asking an AI assistant to retrieve something as sensitive as a driver's license or medical document from my photos. For finding 47 pictures of the same child wearing their favorite blue shirt, though, it could save a lot of scrolling. If you know, you know.
3. Turn a document into a podcast-style conversation
Credit: Future
Gemini can take a document, presentation, Gemini Notebook (formally NotebookLM) or Deep Research report and turn it into an Audio Overview.
Instead of reading a long report, you can listen to two AI hosts discuss the information in a podcast-like conversation. It's especially useful for catching up while driving, walking or working around the house.
ChatGPT Voice can read answers aloud and discuss an uploaded document with you. It does not currently have a dedicated feature that turns the source into this kind of produced, shareable conversation.
Gemini usually takes a few minutes to create an Audio Overview. Once it's finished, you can listen inside the Gemini app, download it or share a link. Google explains how Audio Overviews work here .
This could be useful for more than research papers. I can see myself using it to get through a long school document, meeting notes or a report I need to understand before starting work.
4. Create Personal Gemini Avatars (AI Digital Twin)
Credit: Future
Google allows paid subscribers to build a Personal Avatar, which is a digital clone that looks and sounds like you, powered by Gemini Omni and Google Veo . You can build it right in the Gemini app.
Tap on your profile/settings, and select Personal Avatar(or tap Add Files + Avatarin the prompt window). From there, follow the on-screen prompts on your smartphone to record your face and voice.
You will be asked to turn your head side-to-side and read out lines of text so Gemini can map your facial structure and voice profile.
Once recorded, you can summon yourself in Gemini prompts using @me or @[your username].
You can also create a custom gem with a prompt such as:
"Generate a 3D digital illustration of a friendly male developer with short brown hair and round glasses, wearing a dark blue hoodie, vibrant neon-lit background, studio lighting, high resolution profile avatar."
5. Analyze a YouTube video from a link
Credit: @vidIQ on X
I often find a useful YouTube video and immediately wish I could skip the introduction, sponsorship and repeated explanations. With Gemini, I can paste the link into a chat and ask for the information I actually need.
Gemini can summarize a public YouTube video, answer questions about what was discussed and pull out specific tips or instructions. I could give it a 30-minute product tutorial and ask, "What settings does the presenter recommend changing?" or "List the steps for fixing the problem, but leave out the promotional sections."
Gemini can also find relevant videos, playlists and channels before answering questions about their content. Connecting YouTube lets it personalize some results using your viewing history.
ChatGPT can now do something similar through its Chrome extension, which uses a video's timestamped transcript when captions are available. The difference is where it works. Gemini handles YouTube links inside its regular web and mobile apps, while ChatGPT's officially documented version requires opening the video through Chrome.
Bottom line
ChatGPT remains my choice for plenty of everyday work, but Google is using its existing products to give Gemini abilities that are difficult for another chatbot to copy.
Google Photos search and Android control come from Gemini's position inside Google's ecosystem. Music generation, Audio Overviews and reusable avatars are more surprising because they change what you can create without leaving the chat. Give them a try and let me know what you think in the comments.
