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。

It’s still an early version, but check out how the model handles a challenging puzzle involving both visual and textual clues: (2/3)

Logan Kilpatrick

335,497 subscribers

242,512 次观看 • 1 年前 •via X (Twitter)

教育艺术科学技术

Anya Rossi• Live Now

Private livecam show

12 条评论

Logan Kilpatrick 的头像

Logan Kilpatrick1 年前

Just when you thought it was over... we’re introducing Gemini 2.0 Flash Thinking, a new experimental model that unlocks stronger reasoning capabilities and shows its thoughts. The model plans (with thoughts visible), can solve complex problems with Flash speeds, and more 🧵

Logan Kilpatrick 的头像

Logan Kilpatrick1 年前

Try it out today in Google AI Studio and the Gemini API. This is just the first step in our reasoning journey, excited to see what you all think!

Abduce 的头像

Abduce2 年前

Enjoy a daily word puzzle game with a picture twist! Available now on iOS

Pallav Agarwal 的头像

Pallav Agarwal1 年前

Gemini 2.0 Flash Thinking successfully solves this riddle by answering 'Blue Cone'

adi 的头像

adi1 年前

the fact that it new to flip the 9 is insane wth

idk 的头像

idk1 年前

Give it the owl

TheLobbyistGuy 的头像

TheLobbyistGuy1 年前

Pretty damn interesting.

Sir Mr Meow Meow 的头像

Sir Mr Meow Meow1 年前

very impressive. im curious does it's reasoning chain use other modalities other than text, ie using that super good image gen in the reasoning chains to help figure stuff out, plan ahean, game out various rotations, mark ups, edits using is image & video capabilities?🧐

Nic Polotnianko 🇺🇦 的头像

Nic Polotnianko 🇺🇦1 年前

WTF it's smarter than me WTF

⟁ndrew V 的头像

⟁ndrew V1 年前

thoughts on the 32K context input, is this due to flash level, but with multi modal that context can get consumed quickly I think?

Steven Rianto 的头像

Steven Rianto1 年前

But can it make an ASCII art that write in big letters “HAPPY BIRTHDAY”????

Hrsh 的头像

Hrsh1 年前

More details on Gemini flash 2.0 thinking.

相关视频

How did Abubakar Salim cast @the_ben_starr and @neilnewbon in Dead Take? The new puzzle game from @SurgentStudios offers a "horrific take" on Hollywood, and it even features an ancient headshot from Neil Newbon's early career. Check out the full interview:

Radio Times

13,371 次观看 • 10 个月前

StarVector is out on Hugging Face StarVector is a foundation model for generating Scalable Vector Graphics (SVG) code from images and text. It utilizes a Vision-Language Modeling architecture to understand both visual and textual inputs, enabling high-quality vectorization and text-guided SVG creation.

StarVector is out on Hugging Face StarVector is a foundation model for generating Scalable Vector Graphics (SVG) code from images and text. It utilizes a Vision-Language Modeling architecture to understand both visual and textual inputs, enabling high-quality vectorization and text-guided SVG creation.

AK

254,300 次观看 • 1 年前

This one is the 2013 model. They didn’t upgrade the door handles, but it’s still very hard to tell that it has been upgraded.

This one is the 2013 model. They didn’t upgrade the door handles, but it’s still very hard to tell that it has been upgraded.

EverythingCar_X

328,114 次观看 • 20 天前

Introducing Meta Perception Language Model (PLM): an open & reproducible vision-language model tackling challenging visual tasks. Learn more about how PLM can help the open source community build more capable computer vision systems. Read the research paper, and download the code and dataset:

Introducing Meta Perception Language Model (PLM): an open & reproducible vision-language model tackling challenging visual tasks. Learn more about how PLM can help the open source community build more capable computer vision systems. Read the research paper, and download the code and dataset:

AI at Meta

94,330 次观看 • 1 年前

Meet Qwen-Image-Edit, your smart, seamless image editing companion! Built on the powerful 20B Qwen-Image model, it handles both visual and semantic edits, and delivers state-of-the-art performance. Whether you're refining visuals or transforming styles, it’s got you covered! ✨

Meet Qwen-Image-Edit, your smart, seamless image editing companion! Built on the powerful 20B Qwen-Image model, it handles both visual and semantic edits, and delivers state-of-the-art performance. Whether you're refining visuals or transforming styles, it’s got you covered! ✨

Alibaba Group

64,349 次观看 • 10 个月前

After a long break, I decided to start a company and platform called ComposeFlow, with a mission to empower everyone to create mobile apps with the power of AI and visual development approach🚀 Check out the blog for details: Early version of product:

After a long break, I decided to start a company and platform called ComposeFlow, with a mission to empower everyone to create mobile apps with the power of AI and visual development approach🚀 Check out the blog for details: Early version of product:

Takeshi Hagikura

14,415 次观看 • 1 年前

Git is a must-have skill for developers. But it’s confusing and the learning curve is difficult. Check out this website: Visual and interactive way to learn Git on the web.

Git is a must-have skill for developers. But it’s confusing and the learning curve is difficult. Check out this website: Visual and interactive way to learn Git on the web.

Pratham

309,244 次观看 • 3 年前

It’s 2025 and I’m still buying games off the PS3 store Excited to check out both American McGee’s Alice games this weekend!

It’s 2025 and I’m still buying games off the PS3 store Excited to check out both American McGee’s Alice games this weekend!

Radec

67,613 次观看 • 1 年前

It’s the small details that make the difference (❁´◡`❁) Still not over how cute this model turned out… #Live2D

It’s the small details that make the difference (❁´◡`❁) Still not over how cute this model turned out… #Live2D

𝐌𝐢𝐚 𝐏𝐚𝐩𝐫𝐢𝐤𝐚 | Art&Rig Live2D

30,133 次观看 • 6 个月前

The most insane thing has just happened. Someone dumped an early version of Persona 5 dating half a year before the Japanese release. It contains so much. Early songs, early models, PV02 Last Surprise And this. an early version of Wake Up, Get Up, Get Out There

The most insane thing has just happened. Someone dumped an early version of Persona 5 dating half a year before the Japanese release. It contains so much. Early songs, early models, PV02 Last Surprise And this. an early version of Wake Up, Get Up, Get Out There

Faz

1,531,994 次观看 • 2 年前

brain rotting, but the thoughts still racing // Testing img2vid lipsyncing with the InfiniteTalk model on WaveSpeedAI and a song I made in Suno v5. Love how it handles images with multiple faces and how it animates even when it doesn’t see a mouth. Quick tutorial in 🧵👇 (1/6)

brain rotting, but the thoughts still racing // Testing img2vid lipsyncing with the InfiniteTalk model on WaveSpeedAI and a song I made in Suno v5. Love how it handles images with multiple faces and how it animates even when it doesn’t see a mouth. Quick tutorial in 🧵👇 (1/6)

CoffeeVectors

29,987 次观看 • 8 个月前

SAM 3 tackles a challenging problem in vision: unifying a model architecture for detection and tracking. Christoph, a researcher on SAM 3, shares how the team made it possible. 🔗 Read the SAM 3 research paper:

SAM 3 tackles a challenging problem in vision: unifying a model architecture for detection and tracking. Christoph, a researcher on SAM 3, shares how the team made it possible. 🔗 Read the SAM 3 research paper:

AI at Meta

13,803 次观看 • 8 个月前

Soph. Inf Chase Womack Chase Womack | Grayson Baseball Womack gives Grayson an early lead with this big 2 run 💣. The switch hitter can play all over the infield and handles the barrel well from both sides. @PBRGowins | @PBR_JUCO

Soph. Inf Chase Womack Chase Womack | Grayson Baseball Womack gives Grayson an early lead with this big 2 run 💣. The switch hitter can play all over the infield and handles the barrel well from both sides. @PBRGowins | @PBR_JUCO

Prep Baseball Texas

39,042 次观看 • 1 年前

π0.7 handles diverse prompts that don't just say what to do, but also how to do it, including rich language and multimodal information, such as visual subgoal images. At test time, these images can be produced by a lightweight world model.

π0.7 handles diverse prompts that don't just say what to do, but also how to do it, including rich language and multimodal information, such as visual subgoal images. At test time, these images can be produced by a lightweight world model.

Physical Intelligence

33,306 次观看 • 3 个月前

Excited to release a preview of Moondream 3. A 9B param, 2B active MoE vision language model that makes no compromises; offering state-of-the-art visual reasoning while still retaining an efficient and deployment-friendly form factor.

Excited to release a preview of Moondream 3. A 9B param, 2B active MoE vision language model that makes no compromises; offering state-of-the-art visual reasoning while still retaining an efficient and deployment-friendly form factor.

vik

228,290 次观看 • 10 个月前

Indiana Jones and the Great Circle on Switch 2 is an awesome version of a great game. It’s also full physical on the game card, so let’s open it up and take a look! Thx Bethesda for the early access

Indiana Jones and the Great Circle on Switch 2 is an awesome version of a great game. It’s also full physical on the game card, so let’s open it up and take a look! Thx Bethesda for the early access

Kit & Krysta → Summer Game Fest

15,423 次观看 • 2 个月前