Now that Gemini Embedding 2 is GA, let’s explore... what the model unlocks — from agentic multimodal RAG to visual search — as it maps text, images, video, audio, and documents into a unified embedding space.show more

I vibe coded a visual PDF search app with... ColQwen2. This is how it works: - Store PDF files as images in a Weaviate AI Database vector database - Embed images and text with a multimodal late-interaction model (ColQwen2) - Generate token-wise (and summed) similarity maps to highlight image patches with high similarity Now I need to refactor the messy vibe-coded project. In the meantime, you can check out the Notebook this demo is based on to try it out yourself:show more

Meet Gemma 3n, a model that runs on as... little as 2GB of RAM 🤯 It shares the same architecture as Gemini Nano, and is engineered for incredible performance. We added audio understanding, so now it’s multimodal, fast and lean, and runs on-device (no cloud connection required!)show more

Extract – a system built by the UK government,... using our Gemini foundational model – will help council planners make faster decisions. 🚀 Using multimodal reasoning, it turns complex planning documents – even handwritten notes and blurry maps – into digital data in just 40s. Find out more. ↓show more

Native image generation with Gemini is live 🚀🔥 You... can generate interleaved text and images and edit uploaded images all in a unified model. Big congrats to Robert Riachi Kaushik Shivakumar Fabian Mentzer and many more amazing folks who built this! Try it out and tell us what you think:show more

Gemini 3 Flash is now available in MagicPath ⚡... From our earlier testing, it’s surprisingly strong at visual design for its size, and it’s fast! My recommendation: use it as your first-pass model to explore ideas, then switch to a bigger model for editing + refinements.show more

We're now rolling out interactive images to the Google... Gemini for more visual and dynamic learning. It’s designed to help you visually explore complex academic concepts, turning studying from passive viewing into active exploration. Learn more →show more

Woow Google has just rolled out the AI model... Flash Thinking 2.0 This is the first reasoning model capable of accessing YouTube and it changes everything: - Search for a video on your topic - Ask Gemini to think about the video - You'll have a tailor-made result in 10 sec. And it's even connected to Google Search and Maps!show more

Fable 5 vs Sonnet 5. These evals showcase how... each model prompts Seedance 2.0, the best text-to-video and image-to-video model. In agentic pipelines, that is what sets your success rate per generation.show more

Today, we launched a brand-new intelligent Search box. Here's... what that means: An upgrade to the Search experience with our most advanced Gemini 3.5 models, bringing with them our latest agentic capabilities You can ask across modalities (text, images, files, and videos) and Search can reason across them all We're combining AI Overviews and AI Mode into one, seamless AI Search experience. So you can ask follow-up questions, build context, and received even more tailored and personalized responses This new AI Search experience is live today across desktop and mobile, worldwide.show more

Introducing VL-JEPA: Vision-Language Joint Embedding Predictive Architecture for streaming,... live action recognition, retrieval, VQA, and classification tasks with better performance and higher efficiency than large VLMs. • VL-JEPA is the first non-generative model that can perform general-domain vision-language tasks in real-time, built on a joint embedding predictive architecture. • We demonstrate in controlled experiments that VL-JEPA, trained with latent space embedding prediction, outperforms VLMs that rely on data space token prediction. • We show that VL-JEPA delivers significant efficiency gains over VLMs for online video streaming applications, thanks to its non-autoregressive design and native support for selective decoding. • We highlight that our VL-JEPA model, with an unified model architecture, can effectively handle a wide range of classification, retrieval, and VQA tasks at the same time. by Delong Chen (陈德龙) Mustafa Shukor Théo Moutakanni Willy Jade Lei Yu Tejaswi Kasarla Allen Bolourchi Yann LeCun Pascale Fungshow more

Show-o One Single Transformer to Unify Multimodal Understanding and... Generation discuss: We present a unified transformer, i.e., Show-o, that unifies multimodal understanding and generation. Unlike fully autoregressive models, Show-o unifies autoregressive and (discrete) diffusion modeling to adaptively handle inputs and outputs of various and mixed modalities. The unified model flexibly supports a wide range of vision-language tasks including visual question-answering, text-to-image generation, text-guided inpainting/extrapolation, and mixed-modality generation. Across various benchmarks, it demonstrates comparable or superior performance to existing individual models with an equivalent or larger number of parameters tailored for understanding or generation. This significantly highlights its potential as a next-generation foundation model.show more

Now available: Gemini Pro 1.0 as a base bot!... This multimodal bot accepts text, image, and video input, and has been upgraded with overall quality improvements, including improved reasoning, better latency, and multilingual support. We look forward to seeing what you build!show more

Stop endless app switching. You can now ask Gemini... to complete complex tasks for you across two apps - with one prompt. Look up a location on Maps AND text it to a friend. Ask a question about a video AND save it to notes. What two-in-one tasks will you be asking Gemini?show more

By now, everyone knows that single-vector embedding models are... hugely limiting for modern workflows. But they contain than you think: you can extract sparse Latent Terms from them. And it turns out that BM25 is all you need to turn this vocabulary into a strong retriever.show more

Dreamina Seedance 2.0 is Officially here! Dreamina Seedance 2.0,... ByteDance’s AI-powered creative platform, lets creators easily transform ideas into high-quality videos. You can now edit videos like images, using up to 4 reference modalities (video, image, audio, and text) with precise control over visual effects, camera movements, and more. Why is this revolutionary? - 🔥 One-Prompt Video Editing: Edit videos seamlessly as easily as editing images. - 🎬 Multimodal Creativity: Combine images, video, audio, and text—up to 12 files at once. - 🎥 Remake Viral Content: Create high quality, professional-level videos with intelligent AI. - 🔧 Full Creative Control: Keep consistency in shots, typography, camera flow, and more. This isn’t just another video editing tool, it’s a one-stop AI workspace for all your creative needs. Ready to level up your content? Explore Dreamina Seedance 2.0 and start creating today. 🔗 #dreamina #seedance2 #seedream5 #dreaminatutorial #ai #aitools #aidesign #ecommercedesign #digitalmarketing #startupbusinessshow more

🚨 Baidu Search has launched its most significant upgrade... in a decade - and it's a total game-changer. 🔥 Here’s why it’s a big deal in the evolution of search technology: • The new Smart Search Box handles more complex queries—over 1,000 Chinese characters—and lets you search using images, voice, video, or files. • Baidu’s AI Assistant now supports video calls, multimodal input (text, voice, image, files), and richer outputs like notes, images, and videos. Its enhanced Deep Search capability mimics human reasoning—perceiving, planning, executing, and generating—providing multi-layered insights beyond traditional search. Search is now smarter, more useful, and more intuitive than ever! ⚡🌟show more

Resolv is here to stablecoinize everything. The next phase... live: clusters. A modular way to integrate productive assets across DeFi and beyond, embedding their yields directly into stablecoin collateral. It begins with Fluid 🌊, where lending flows now power the stable rails.show more

That dog in your photo? He's got something to... say. 🐶 Turn your images into eight-second video clips with sound effects and speech in the Google Gemini and Flow from Google Labs. This feature uses Veo 3 to generate motion that reflects real world physics and includes a new experimental audio capability so you can really bring your images to life. Try it at andshow more

This is WILD. Freepik just gave AI video a... voice with Wan 2.5. Smooth motion. Clean visuals. And now? Built-in audio from just text or images. It’s fast, fluid, and free to try. Here’s what you can generate in seconds:show more

AI agents can already read the web. Now we're... giving them eyes. Introducing Semantic Video Search, powered by the UpRock Network. Not transcript search. Not keyword matching. Real multimodal intelligence across: → video → audio → frames → timestamps → context Ask questions. Get answers from video. Powered by a global swarm of 3M+ real devices across 190+ countries, UpRock gives AI agents access to the internet as real users experience it. No datacenter blocks. No proxy games. No synthetic traffic. Just ground-truth intelligence from the real web. Over the next few posts, we'll show what this unlocks. 👇show more