Lijun Yu retweetledi
We’re launching Nano Banana 2, built on the latest Gemini Flash model. 🍌
It’s state-of-the-art for creating and editing images, combining Pro-level capabilities with lightning-fast speed. 🧵
GIF
English
Lijun Yu
38 posts
Lijun Yu
@LijunYu0
Research Scientist @GoogleDeepMind | AI PhD @CarnegieMellon | CS/Econ @PKU1898
Mountain View, CA Katılım Ağustos 2014
108 Takip Edilen595 Takipçiler
Lijun Yu retweetledi
We’ve upgraded our specialized reasoning mode Gemini 3 Deep Think to help solve modern science, research, and engineering challenges – pushing the frontier of intelligence. 🧠
Watch how the Wang Lab at Duke University is using it to design new semiconductor materials. 🧵
English
Lijun Yu retweetledi
How could AI act as a better research collaborator? 🧑🔬
In two new papers with @GoogleResearch, we show how Gemini Deep Think uses agentic workflows to help solve research-level problems in mathematics, physics, and computer science.
More → goo.gle/4aGs3Pz
English
Lijun Yu retweetledi
We’re pushing the boundaries of intelligence even further with Gemini 3 Deep Think. 🧠
This mode meaningfully improves reasoning capabilities by exploring many hypotheses simultaneously to solve problems.
Here’s how it coded a simulated dominoes game from a single prompt ⬇️
English
Lijun Yu retweetledi
Gemini 3 Deep Think mode is now live in the Gemini app for Ultra users. 🚀
Building on the technology that reached a gold-medal level at the ICPC World Finals & IMO, it uses parallel thinking to excel at difficult coding and scientific tasks. bit.ly/48EPzdz
English
Check out our new paper on the future of pixel space generative pretraining, and the insufficiency of current compute! This is one more evidence for the ever growing compute need, and by 2030 we can make this happen :)
Xinchen Yan@skywalkeryxc
For years, RAW pixel space pretraining has been sidelined: too compute-expensive. Our new @GoogleDeepMind paper 📜 dives into the scaling trends of raw pixel models to answer the question “how far are we from scaling up next-pixel prediction?” arxiv.org/pdf/2511.08704 Forecast: Raw next-pixel modeling will reach competitive ImageNet classification (>80% top1 accuracy) and generation metrics (90 Fr’echet Distance) in five years! Threads 👇
English
Meet Nano Banana Pro 🍌🍌🍌, bringing image generation and editing to a whole new level with Gemini 3.
Google DeepMind@GoogleDeepMind
We just dropped Nano Banana Pro, built on Gemini 3. 🍌 With state-of-the-art text rendering, vast world knowledge and studio-quality creative controls, Gemini 3 Pro Image can create and edit more complex visuals, infographics and more. Here’s what’s under the hood. 🧵
English
Lijun Yu retweetledi
🚨BREAKING: @GoogleDeepMind’s Gemini-3-Pro is now #1 across all major Arena leaderboards
🥇#1 in Text, Vision, and WebDev - surpassing Grok-4.1, Claude-4.5, and GPT-5
🥇#1 in Coding, Math, Creative Writing, Long Queries, and nearly all occupational leaderboards.
Massive gains over Gemini-2.5:
🔸WebDev in Code Arena: 1487 (+280 pts vs 2.5)
🔸Text: 1501 (+50 pts)
🔸Vision: 1328 (+70 pts)
🔸Arena Expert: Top-3 (just 3 pts behind #1)
Huge congrats to the @GoogleDeepMind team on this breakthrough! 👏
Sundar Pichai@sundarpichai
Introducing Gemini 3 ✨ It’s the best model in the world for multimodal understanding, and our most powerful agentic + vibe coding model yet. Gemini 3 can bring any idea to life, quickly grasping context and intent so you can get what you need with less prompting. Find Gemini 3 Pro rolling out today in the @Geminiapp and AI Mode in Search. For developers, build with it now in @GoogleAIStudio and Vertex AI. Excited for you to try it!
English
Feel the power of our most intelligent model, Gemini 3 🚀🚀🚀
Google DeepMind@GoogleDeepMind
This is Gemini 3: our most intelligent model that helps you learn, build and plan anything. It comes with state-of-the-art reasoning capabilities, world-leading multimodal understanding, and enables new agentic coding experiences. 🧵
English
Lijun Yu retweetledi
🧵 Why Veo 3.1 ranks #1 on the Video Arena, and Sora 2 went viral.
Sora 2 by @OpenAI captured massive attention, but the community has shown us that @GoogleDeepMind's Veo 3.1 leads on core model quality. The difference comes down to two standout features that fueled Sora 2’s rise:
🧑🤝🧑 Cameos & 📚 Narratives
Let's dive into these in the thread 👇
Arena.ai@arena
🚨🎬 Big news from Video Arena! @GoogleDeepMind’s latest Veo 3.1 now ranks #1 in both Text-to-Video and Image-to-Video leaderboards. 🏆 This is a +30-point leap from Veo 3.0 → 3.1, making it the first model to break 1400 in Video Arena history! Huge congrats to the @GoogleDeepMind team for pushing the frontier of video generation forward! More details in the thread 🧵
English
Excited to share that Veo 3.1 is winning across the board 🏆
Arena.ai@arena
🚨🎬 Big news from Video Arena! @GoogleDeepMind’s latest Veo 3.1 now ranks #1 in both Text-to-Video and Image-to-Video leaderboards. 🏆 This is a +30-point leap from Veo 3.0 → 3.1, making it the first model to break 1400 in Video Arena history! Huge congrats to the @GoogleDeepMind team for pushing the frontier of video generation forward! More details in the thread 🧵
English
Lijun Yu retweetledi
Veo 3.1 makes a big impact on the small details of your video generations. See how different surfaces, fabrics, and textures feel even truer to life with this new model. 🧵
English
Lijun Yu retweetledi
build with Veo 3.1 on the Gemini API, now with:
• richer audio
• enhanced realism
• stronger prompt adherence
• new control capabilities
English
Lijun Yu retweetledi
Today we are introducing Veo 3.1, our newest video generation model. Here's what that means for Flow:
1. Your videos will have richer audio/dialogue, more narrative control, enhanced realism, and more.
2. It's available across Flow's current suite of features, including Frames to Video, Ingredients to Video, and Extend.
Here's how you can use each feature to level up your AI filmmaking:
English
Thrilled to introduce Veo 3.1🚀, where I am fortunate to have led a great group of people to enable online reinforcement learning with thinking for video generation at scale for the first time and achieved state-of-the-art across tasks.
Google DeepMind@GoogleDeepMind
Veo is getting a major upgrade. 🚀 We’re rolling out Veo 3.1, our updated video generation model, alongside improved creative controls for filmmakers, storytellers, and developers - many of them with audio. 🧵
English
ICPC Gold with Gemini Deep Think!
Quoc Le@quocleix
(1/3) Thrilled to announce a new Gemini breakthrough! Building on our success at IMO this year, an advanced version of Gemini Deep Think achieved gold-medal level performance at the ICPC 2025 World Finals - one of the world’s leading competitive programming competitions. deepmind.google/discover/blog/…
English
Now onto endless Deep Think with IMO gold-medal!
Quoc Le@quocleix
Following its IMO gold-level win, @GoogleDeepMind is sharing Gemini Deep Think with mathematicians for feedback. Excited to see what they discover! 🧠 Plus, an updated Gemini 2.5 Deep Think is now rolling out for Google AI Ultra subscribers. Learn more: bit.ly/3IWcWq0
English
Excited to share #TexTok at #CVPR today! Join us for a journey on image tokenization at Oral 4A and Poster 252.
Kaiwen Zha@KaiwenZha
Excited to present our #CVPR2025 oral paper #TexTok @CVPR! #TexTok is an image tokenizer that uses text during tokenization, achieving both high recon/gen quality and low cost. 🎙️Oral: Sat, June 14, 1:15-1:30PM CDT (Session 4A) 📌Poster: Sat, June 14, 5-7PM CDT (ExHall D #252)
English
Lijun Yu retweetledi
1/n: 🚀 Announcing YuE (乐) – the most powerful open-source full-song music generation model! 🎵 Tackle the lyrics-to-song task (like Suno.ai) with support for diverse genres, stunning vocals, & multiple languages. Bonus? It’s Hugging Face & LLAMA-compatible for easy fine-tuning. 🛠️
Code: github.com/multimodal-art…
Demo: map-yue.github.io
English