正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Also, we're exploring new ways to create games using Gemini 3 - we've created and published a few experimental YouTube Playables. Each of the games shown in the video below was created by Gemini 3 from scratch w/a few natural language prompts/images. Try them out at 👾👾

Jeff Dean

521,998 subscribers

776,746 次观看 • 10 个月前 •via X (Twitter)

游戏科学技术

Anya Rossi• Live Now

Private livecam show

37 条评论

Logan Kilpatrick 的头像

Logan Kilpatrick10 个月前

back to the DeepMind roots :)

bone 的头像

bone10 个月前

Can we vibecode youtube playables and earn ad money?

Petr Brzek 的头像

Petr Brzek10 个月前

Not in EUROPE 🥲

Houston Austin Muzamhindo 的头像

Houston Austin Muzamhindo10 个月前

Welcome back. We missed you.

dan 的头像

dan10 个月前

we're very clearly in the flash games era of ai games and it's great

Delip Rao e/σ 的头像

Delip Rao e/σ10 个月前

so fun to vibe code on 3-Pro.

Jos van der Westhuizen 的头像

Jos van der Westhuizen10 个月前

Gemini 3 is also really good at handling game engine tasks! I one-shot created this basic Unity game

hengcherkeng 的头像

hengcherkeng10 个月前

Kaggle has a 2d shape packing optimization competition running now. Someone can try Gemini 3 to run a simulation/game on packing n trees into the smallest square. Maybe we can see surprising results?

Jeff Dean 的头像

Jeff Dean10 个月前

Tree search!

hengcherkeng 的头像

hengcherkeng10 个月前

Gemini 3 models is very good on ARC-AGI-2 (2X SOTA jump to 31.11%, 45.14%). I think Gemini 3 is good at "tree search" and optimization. I will work on it!

Jason Mayes 的头像

Jason Mayes10 个月前

Wow it actually works first try for vanilla JavaScript just as I wanted - what magic is this?! As its web link to @CodePen to try live:

Ben 的头像

Ben10 个月前

I've created this with Gemini 3 already - new gaming era is here

Mtu Nai 的头像

Mtu Nai10 个月前

(Sad face) :-( 😔

Jason Mayes 的头像

Jason Mayes10 个月前

Antigravity working well to vibe code this audio vis into existence:

FARHAN AKBAR 的头像

FARHAN AKBAR10 个月前

Wild to see Gemini 3 go from idea → fully playable game with just a few prompts. This isn’t just game creation it’s a new creative medium. Can’t wait to see how far this goes once the community starts experimenting.

Jason Mayes 的头像

Jason Mayes10 个月前

This is brilliant!

Sam 的头像

Sam10 个月前

holy shit, these are soo addictive!

j r j 007007 的头像

j r j 00700710 个月前

give us the prompts!

JOHN Z⚡ 的头像

JOHN Z⚡10 个月前

I’m honestly blown away by this model, seriously! It’s a really powerful one, huge and impressive in every way. Google absolutely nailed it!

Karl Tuyls 的头像

Karl Tuyls10 个月前

Very cool, congrats!

Muratcan Koylan 的头像

Muratcan Koylan10 个月前

omg they are amazing

Readable AI 的头像

Readable AI10 个月前

Ahhh! Single-dev games are sooo back. Can’t wait for the “made by one person in a caffeine-induced trance” era to return. 💀

烟花老师的头像

烟花老师10 个月前

that's my cases, Gemini3 is awesome!

Lou 的头像

Lou10 个月前

Is there a way to share the prompts that went into this or a live session in the future?

Jason Mayes 的头像

Jason Mayes10 个月前

Live CodePen demo here: - front end JS ftw:

Alex 的头像

Alex10 个月前

@ashayasharma 😭

Yashjit 的头像

Yashjit10 个月前

Is there any future possibilities of making our own games and publishing as playables on YouTube?

Rituraj 的头像

Rituraj10 个月前

This is wild — Gemini 3 is basically speed-running game development at this point. From prompts to a playable prototype in minutes? The future is getting fun 👾🔥

Garrett 的头像

Garrett10 个月前

more games!

Sourav Boyal 的头像

Sourav Boyal10 个月前

No Flash 3 this time?

Ukraine Mentioned 🇺🇦 的头像

Ukraine Mentioned 🇺🇦10 个月前

this is pathetic af, no game works

Jos van der Westhuizen 的头像

Jos van der Westhuizen10 个月前

Been testing Gemini3.0 a lot for our game creation in Unity and it's really good!

Stern - uɹǝʇS 的头像

Stern - uɹǝʇS10 个月前

What is the limitation for playables on YouTube? If i can upload a game, is it feasible to host other Gemini apps on YouTube? Could we create custom media players?

Axel Rios 的头像

Axel Rios10 个月前

wow, love this

Himanshu Kumar 的头像

Himanshu Kumar10 个月前

Wow, that's impressive, Jeff, but I wonder if these AI games will truly offer the same fun, you know?

Tim Shi 的头像

Tim Shi10 个月前

Very impressive. Congrats!

Tim Kostolansky 的头像

Tim Kostolansky10 个月前

do ppl want these?

相关视频

We used Gemini 3 to create bite-sized demo games on YouTube Playables with just a few simple prompts 🎮 Give them a try →

We used Gemini 3 to create bite-sized demo games on YouTube Playables with just a few simple prompts 🎮 Give them a try →

Google DeepMind

116,471 次观看 • 10 个月前

Paper Mario’s partners now unfold from his pocket like in the games! A few of the partner abilities are also shown, so check out the full video!

Paper Mario’s partners now unfold from his pocket like in the games! A few of the partner abilities are also shown, so check out the full video!

mRama

86,608 次观看 • 10 个月前

We're going live in one hour! Tune in for a hands-on look at Gemini 3, Nano Banana Pro, Veo and how to create a full brand ecosystem from scratch. →

Google Cloud Tech

12,024 次观看 • 9 个月前

GEMINI 3 LAUNCH IS HERE I got a SNEAK PEEK at Gemini 3 with Logan Kilpatrick (Google Deepmind), and it might be the most POWERFUL vibe-coding tool on the planet. A little breakdown: 1. Anyone can build 3D and casual games now You can vibecode full, playable 3D video games generated in minutes. Actual games with physics, characters, controls, and loops you can remix instantly. Pure insanity. I can see founders and brands spinning up games on the fly to ride trends and drive growth. 2. Intelligent apps are becoming the default We built apps where reasoning, memory, and multi-step planning were baked in from the start. Once you’re building apps with ACTUAL intelligence baked in, there’s a whole wave of new opportunities that weren’t possible before. 3. Gemini acts like a creative partner You describe the idea, Gemini fills in the gaps, challenges decisions, proposes alternatives, and iterates in real time. 4. Vibe coding hits a new level Gemini 3 can generate assets, code, game logic, UI, and narrative in one flow. Tools like Claude and Cursor feel fast. This feels like the next layer, the one where a single builder can compete with full teams. Logan Kilpatrick and I pushed Google Gemini 3 hard, and the outputs were solid. A few times we had to give it a few extra prompts but it took feedback really well. I think 1 year ago, a lot of people discounted Google in the AI arms race. Can you discount them anymore? Doubt it. After this, it feels like they at best leading, at worst leading. What do you think of Google's AI efforts/Gemini 3 My biggest takeaway was how it just felt like Gemini 3 had a little more vibe coding horsepower than anything I’ve used.

GEMINI 3 LAUNCH IS HERE I got a SNEAK PEEK at Gemini 3 with Logan Kilpatrick (Google Deepmind), and it might be the most POWERFUL vibe-coding tool on the planet. A little breakdown: 1. Anyone can build 3D and casual games now You can vibecode full, playable 3D video games generated in minutes. Actual games with physics, characters, controls, and loops you can remix instantly. Pure insanity. I can see founders and brands spinning up games on the fly to ride trends and drive growth. 2. Intelligent apps are becoming the default We built apps where reasoning, memory, and multi-step planning were baked in from the start. Once you’re building apps with ACTUAL intelligence baked in, there’s a whole wave of new opportunities that weren’t possible before. 3. Gemini acts like a creative partner You describe the idea, Gemini fills in the gaps, challenges decisions, proposes alternatives, and iterates in real time. 4. Vibe coding hits a new level Gemini 3 can generate assets, code, game logic, UI, and narrative in one flow. Tools like Claude and Cursor feel fast. This feels like the next layer, the one where a single builder can compete with full teams. Logan Kilpatrick and I pushed Google Gemini 3 hard, and the outputs were solid. A few times we had to give it a few extra prompts but it took feedback really well. I think 1 year ago, a lot of people discounted Google in the AI arms race. Can you discount them anymore? Doubt it. After this, it feels like they at best leading, at worst leading. What do you think of Google's AI efforts/Gemini 3 My biggest takeaway was how it just felt like Gemini 3 had a little more vibe coding horsepower than anything I’ve used.

GREG ISENBERG

73,886 次观看 • 10 个月前

Also starting today, Gemini Advanced users can create with Canvas on Gemini 2.5 Pro. This model’s advanced coding abilities make it easy for anyone to collaborate, create, and share imaginative prototypes (like this example of a 3D dino world, made in under 5 minutes with just a few prompts). Here’s how to get started: 1) Go to 2) Choose the “2.5 Pro (experimental)” model 3) Select “Canvas” and start prompting We’re working hard to make Gemini 2.5 Pro available for more people to try soon–watch this space for updates.

Also starting today, Gemini Advanced users can create with Canvas on Gemini 2.5 Pro. This model’s advanced coding abilities make it easy for anyone to collaborate, create, and share imaginative prototypes (like this example of a 3D dino world, made in under 5 minutes with just a few prompts). Here’s how to get started: 1) Go to 2) Choose the “2.5 Pro (experimental)” model 3) Select “Canvas” and start prompting We’re working hard to make Gemini 2.5 Pro available for more people to try soon–watch this space for updates.

Google Gemini

302,724 次观看 • 1 年前

We asked AI Mode in Search how a basketball player's 3 point shot relates to the quadratic equation. With Gemini 3's new generative UI capabilities, it created an interactive visualization to help bring this concept to life. Try it out with the Gemini 3 drop down in AI Mode and share your examples with us!

We asked AI Mode in Search how a basketball player's 3 point shot relates to the quadratic equation. With Gemini 3's new generative UI capabilities, it created an interactive visualization to help bring this concept to life. Try it out with the Gemini 3 drop down in AI Mode and share your examples with us!

Rajan Patel

202,459 次观看 • 10 个月前

You can now try Gemini 3 Pro in Figma Make! What stands out to me from early testing is how useful Gemini 3 is for exploring visual directions + ideas that designers can then shape and push further with their craft. My thoughts (and some 🤯 demos) in the video below...

You can now try Gemini 3 Pro in Figma Make! What stands out to me from early testing is how useful Gemini 3 is for exploring visual directions + ideas that designers can then shape and push further with their craft. My thoughts (and some 🤯 demos) in the video below...

Dylan Field

116,382 次观看 • 10 个月前

I made a sprite animator using Google's Gemini 2.0 flash experimental. Here's how it works: 1. Gemini 2.0 exp generates the images 2. They are uploaded each in an S3 bucket/folder 3. Then you can use WebGPU /w canvas or just img Let's watch some examples and generate some:

I made a sprite animator using Google's Gemini 2.0 flash experimental. Here's how it works: 1. Gemini 2.0 exp generates the images 2. They are uploaded each in an S3 bucket/folder 3. Then you can use WebGPU /w canvas or just img Let's watch some examples and generate some:

wavefnx

353,002 次观看 • 1 年前

Come check out the Akupara Games 10 Year Anniversary Direct on Twitch today! It takes place today at 12pm PT/3pm ET (3 hrs from now) and the show was created by and is presented by me! Here's to another 10 years unique indie games from Akupara Games 💫 Indie Games!

Come check out the Akupara Games 10 Year Anniversary Direct on Twitch today! It takes place today at 12pm PT/3pm ET (3 hrs from now) and the show was created by and is presented by me! Here's to another 10 years unique indie games from Akupara Games 💫 Indie Games!

Alpha Beta Gamer

13,067 次观看 • 20 天前

A few environments, created locally on my PC in a few seconds each using that cool new Apple image to Gaussian splat tech, stitched together and lit in Octane Render 2026 and combined with a few models made using Hyper3D by Deemos Rodin. Next year is bound to be the year of Generative 3d.

A few environments, created locally on my PC in a few seconds each using that cool new Apple image to Gaussian splat tech, stitched together and lit in Octane Render 2026 and combined with a few models made using Hyper3D by Deemos Rodin. Next year is bound to be the year of Generative 3d.

Martin Nebelong

10,485 次观看 • 9 个月前

This New Tutorial Video Will Teach You How To Create A Complete Professional AI Movie from Scratch for Free, Using 3 Prompts on Google Flow, For Youtube Automation. The Entire Prompt is on the video Description.

This New Tutorial Video Will Teach You How To Create A Complete Professional AI Movie from Scratch for Free, Using 3 Prompts on Google Flow, For Youtube Automation. The Entire Prompt is on the video Description.

BIG CHRIS

28,382 次观看 • 4 天前

Gemini in Chrome is unbeatable for learning new things You can turn a YouTube video into a full tutorial and ask follow-up questions. - Copy the video URL - Type “@ Gemini” in the URL bar and your prompt - Paste the video URL at the end The full tutorial is created in 2 sec!

Gemini in Chrome is unbeatable for learning new things You can turn a YouTube video into a full tutorial and ask follow-up questions. - Copy the video URL - Type “@ Gemini” in the URL bar and your prompt - Paste the video URL at the end The full tutorial is created in 2 sec!

Paul Couvert

310,608 次观看 • 1 年前

Tested out the new Kling AI 2.0 model on Sunday afternoon. Using ChatGPT, I created a bunch of challenging prompts for images in Midjourney and then animated them in Kling 2.0. It's an incredible leap forward for AI motion. I'll comment on some of the tests below.

Tested out the new Kling AI 2.0 model on Sunday afternoon. Using ChatGPT, I created a bunch of challenging prompts for images in Midjourney and then animated them in Kling 2.0. It's an incredible leap forward for AI motion. I'll comment on some of the tests below.

Jason Zada

52,911 次观看 • 1 年前

This is Project Requiem 📽️ An AI Short Film Created entirely in Veo 3 using Flow by Google Prompted in JSON, YAML, and natural language One AI VFX shot was created with Runway Aleph This film explores the struggles of separating reality from simulated reality

This is Project Requiem 📽️ An AI Short Film Created entirely in Veo 3 using Flow by Google Prompted in JSON, YAML, and natural language One AI VFX shot was created with Runway Aleph This film explores the struggles of separating reality from simulated reality

Dave Clark

65,118 次观看 • 1 年前

We're bringing more fun to the games you love across all the devices you play on. 🎉 A supercharged gaming profile, in-game help from Gemini, and a new personalized hub for it all. And that’s just the start. Take a deep dive into a new era of Google Play Games. #GoPlay

Google Play

25,414 次观看 • 1 年前

Grok Imagine has an excellent image editor. The video below was created from a single image (created by Grok Imagine) that I modified using Grok Imagine's image editor. If you have any trouble using it, let me know in the comments below. I'll try to help you.

Grok Imagine has an excellent image editor. The video below was created from a single image (created by Grok Imagine) that I modified using Grok Imagine's image editor. If you have any trouble using it, let me know in the comments below. I'll try to help you.

Déborah

1,824,230 次观看 • 8 个月前