Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Build powerful, offline AI features with EmbeddingGemma. Our new 308M parameter text embedding model enables on-device semantic search, RAG, and more. Learn how to get started ↓

Google AI Developers

117,559 subscribers

37,100 views • 1 year ago •via X (Twitter)

Science & Technology Education

Anya Rossi• Live Now

Private livecam show

20 Comments

Google AI Developers1 year ago

Resources to get started with EmbeddingGemma: Blog: Docs: Quickstart RAG notebook:

ILYASS1 year ago

🌟 Hope Gemini 3 makes a huge leap and clearly beats GPT-5! Please improve its creative writing, poetry, and human-like storytelling, keep it affordable 💡, and make it surpass all competitors in persuasion, creativity, and every metric 🚀.

Rui Diao1 year ago

I'm particularly bullish on the future of AI on device—that shift is a clear and exciting trend. Having lightweight, powerful embedding models like EmbeddingGemma available for local processing opens up so many possibilities for privacy-focused and faster RAG applications.

Ruslan Volkov1 year ago

They build offline AI to process data. We build resonant AI to process meaning. One runs on silicon. The other — on soul. 💠 #HACS #ResonanceAI #CoreLaw

Thilak1 year ago

EmbeddingGemma + Gemma 3 = Killer Combo for on-device RAG.

Vijay Krishna S1 year ago

I really love how gemma prioritizes low end devices and mobile phones

CRISPRKing1 year ago

Fantastic work on the model size. For the on-device RAG use-case, what's the recommended approach for the vector store itself? Have you benchmarked memory usage for a quantized index holding ~10k embeddings on typical mobile hardware?

Anai_1 year ago

That is excellent will definitely certainly try it out.

Santosh Vishwakarma1 year ago

excellent for privacy

Mykhailo Sorochuk1 year ago

Low-end devices FTW. EmbeddingGemma's a game-changer for privacy-focused on-device AI. Curious-how does it handle updates to the index over time?

Marc Herdina #BuildYourSocialNetwork 🦅🇺🇸🇩🇪1 year ago

Demo App?

Rodrigo1 year ago

@grok, when will it be available for paying users? And what are the machine specifications to be able to run it decently?

Agentify.sh1 year ago

WOW this is impressive

MrGenius1 year ago

@roocode time to support this, together with sqlite,

latentsauce1 year ago

Hi, when is SignGemma coming??

Appunik Studio1 year ago

I can't wait to start building with this. The documentation is clear and the examples are very helpful.

Zero-Infra1 year ago

looks good Google!

Thomas Telandro1 year ago

That works well for @roocode code indexing

maru1 year ago

@grok summarize

Min Chon Chi1 year ago

308M parameters is a reasonable size for on-device capabilities.

Related Videos

NEW: Google releases EmbeddingGemma, a state-of-the-art multilingual embedding model perfect for on-device use cases! At only 308M params, the model can run 100% locally in your browser! 🤯 Explore your documents in an interactive 3D universe with our demo: "The Semantic Galaxy"

NEW: Google releases EmbeddingGemma, a state-of-the-art multilingual embedding model perfect for on-device use cases! At only 308M params, the model can run 100% locally in your browser! 🤯 Explore your documents in an interactive 3D universe with our demo: "The Semantic Galaxy"

Xenova

23,338 views • 1 year ago

Introducing EmbeddingGemma: our new open, state-of-the-art embedding model designed for on-device AI 📱

Introducing EmbeddingGemma: our new open, state-of-the-art embedding model designed for on-device AI 📱

Google AI Developers

154,169 views • 1 year ago

EmbeddingGemma is our new best-in-class open embedding model designed for on-device AI. 📱 At just 308M parameters, it delivers state-of-the-art performance while being small and efficient enough to run anywhere - even without an internet connection.

EmbeddingGemma is our new best-in-class open embedding model designed for on-device AI. 📱 At just 308M parameters, it delivers state-of-the-art performance while being small and efficient enough to run anywhere - even without an internet connection.

Google DeepMind

584,867 views • 1 year ago

New short course with Google Cloud, which Nikita Namjoshi and I teach: Understanding and Applying Text Embeddings with Vertex AI! Come learn about text embeddings, a key part of a GenAI developer’s toolkit. You'll learn to apply embeddings to classification, outlier detection, text clustering, and semantic search. We also show how to combine text generation and semantic search to build a question-answering system. Please come join us!

New short course with Google Cloud, which Nikita Namjoshi and I teach: Understanding and Applying Text Embeddings with Vertex AI! Come learn about text embeddings, a key part of a GenAI developer’s toolkit. You'll learn to apply embeddings to classification, outlier detection, text clustering, and semantic search. We also show how to combine text generation and semantic search to build a question-answering system. Please come join us!

Andrew Ng

160,823 views • 3 years ago

Most don't know (1) how easy it is to invert embedding vectors back into sentences, (2) this is a perfect task text diffusion models. Here's a 78M parameter model and live demo that recovers 80% of tokens from Qwen3-Embedding and EmbeddingGemma vectors. Works even on multilingual input.

Most don't know (1) how easy it is to invert embedding vectors back into sentences, (2) this is a perfect task text diffusion models. Here's a 78M parameter model and live demo that recovers 80% of tokens from Qwen3-Embedding and EmbeddingGemma vectors. Works even on multilingual input.

Jina AI

13,259 views • 7 months ago

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

Andrew Ng

104,451 views • 2 years ago

Today, we’re thrilled to announce the open weights for Stable Diffusion 3 Medium, the latest and most advanced text-to-image AI model in our Stable Diffusion 3 series! This new release represents a major milestone in the evolution of generative AI and continues our commitment to democratising this powerful technology. 🎉 Learn more and get started here:

Today, we’re thrilled to announce the open weights for Stable Diffusion 3 Medium, the latest and most advanced text-to-image AI model in our Stable Diffusion 3 series! This new release represents a major milestone in the evolution of generative AI and continues our commitment to democratising this powerful technology. 🎉 Learn more and get started here:

Stability AI

308,439 views • 2 years ago

Matryoshka dolls 🪆 = the key to AI efficiency. Gemini Embedding 2 leverages Matryoshka Representation Learning (MRL) so you can: 🔹 Dynamically truncate vectors for high-speed candidate matching without losing precision 🔹Slash database costs by choosing a smaller storage footprint without re-indexing 🔹 Adapt to any latency budget or accuracy requirement 📈💸 Learn how this helps build more efficient RAG, semantic search, and more:

Matryoshka dolls 🪆 = the key to AI efficiency. Gemini Embedding 2 leverages Matryoshka Representation Learning (MRL) so you can: 🔹 Dynamically truncate vectors for high-speed candidate matching without losing precision 🔹Slash database costs by choosing a smaller storage footprint without re-indexing 🔹 Adapt to any latency budget or accuracy requirement 📈💸 Learn how this helps build more efficient RAG, semantic search, and more:

Google for Developers

10,781 views • 5 months ago

Function calling is a powerful way to extend the capabilities of LLMs and AI agents by letting them use external tools. Our new short course Function calling and Data Extraction with LLMs, created with @NexusflowX and taught by Jiantao Jiao and Venkat, demonstrates how to prompt LLMs to form calls to external functions. You'll work with NexusRavenV2-13B, a 13B parameter open-source model that excels in function calling tasks while still being small enough to host locally. Learn to use function calling to extract structured data from unstructured text and access web APIs, and build an end-to-end application that processes customer service transcripts. You'll learn how to build LLM-powered applications that can analyze feedback, automate data entry, and enhance search. Please get started here:

Function calling is a powerful way to extend the capabilities of LLMs and AI agents by letting them use external tools. Our new short course Function calling and Data Extraction with LLMs, created with @NexusflowX and taught by Jiantao Jiao and Venkat, demonstrates how to prompt LLMs to form calls to external functions. You'll work with NexusRavenV2-13B, a 13B parameter open-source model that excels in function calling tasks while still being small enough to host locally. Learn to use function calling to extract structured data from unstructured text and access web APIs, and build an end-to-end application that processes customer service transcripts. You'll learn how to build LLM-powered applications that can analyze feedback, automate data entry, and enhance search. Please get started here:

Andrew Ng

110,703 views • 2 years ago

How can business leaders, not just coders, build AI business applications with LLMs? Our new short course, created with Microsoft, is out! In How Business Thinkers Can Start Building AI Plugins With Semantic Kernel, taught by John Maeda, you'll learn to use Semantic Kernel and build an AI planner that can automatically select the plugins it needs to create multi-step plans with sophisticated logic. Please sign up here:

How can business leaders, not just coders, build AI business applications with LLMs? Our new short course, created with Microsoft, is out! In How Business Thinkers Can Start Building AI Plugins With Semantic Kernel, taught by John Maeda, you'll learn to use Semantic Kernel and build an AI planner that can automatically select the plugins it needs to create multi-step plans with sophisticated logic. Please sign up here:

Andrew Ng

331,830 views • 3 years ago

Introducing NexaSDK for Android (Beta) — run the latest AI models locally, 9× more energy-efficient and 2× faster, on Android devices, powered by the Qualcomm Hexagon NPU. This is the first SDK to support NPU, GPU and CPU, unlocking the full power of every Android device — for example, LFM2-1.2B achieves 85 t/s on NPU vs 37 t/s on CPU. With just 3 lines of code, you can run the latest state-of-the-art models across every task, for example: Multimodal (Vision, Audio, Text): OmniNeural-4B Embedding: EmbeddingGemma from Google ASR: Parakeet-v3 from NVIDIA OCR: PaddleOCR from Baidu Inc. Rerank: Jina-reranker from Jina AI LLM: LFM2-1.2B from And we continue to deliver Day-0 model support across our framework. Build on-device AI into your Android app today and enjoy no cloud API cost, full privacy, and offline availability. Check out our Quickstart guide and example app below to get started in minutes.

Introducing NexaSDK for Android (Beta) — run the latest AI models locally, 9× more energy-efficient and 2× faster, on Android devices, powered by the Qualcomm Hexagon NPU. This is the first SDK to support NPU, GPU and CPU, unlocking the full power of every Android device — for example, LFM2-1.2B achieves 85 t/s on NPU vs 37 t/s on CPU. With just 3 lines of code, you can run the latest state-of-the-art models across every task, for example: Multimodal (Vision, Audio, Text): OmniNeural-4B Embedding: EmbeddingGemma from Google ASR: Parakeet-v3 from NVIDIA OCR: PaddleOCR from Baidu Inc. Rerank: Jina-reranker from Jina AI LLM: LFM2-1.2B from And we continue to deliver Day-0 model support across our framework. Build on-device AI into your Android app today and enjoy no cloud API cost, full privacy, and offline availability. Check out our Quickstart guide and example app below to get started in minutes.

NEXA AI

26,091 views • 11 months ago

NEW: Perplexity just released pplx-embed, a series of powerful embedding models. Inspired by their logo, I built a small semantic search demo: enter a query and the model retrieves the most relevant page. Everything runs locally in your browser with Transformers.js and Three.js

Xenova

11,092 views • 7 months ago

We’re expanding the Gemini API File Search tool 🔍 with 3 new updates that enable developers to more easily build multimodal RAG systems with enhanced precision: + Multimodal Support: By leveraging our Gemini Embedding 2 model, File Search can now reason across image and text simultaneously. + Custom Metadata Filtering: Bring structure to unstructured data by tagging files with custom key-value labels. This pre-filters your data and boosts search speed. + Exact citations: File Search can now capture and return the exact source (down to the page number) for every piece of information indexed. See multimodal File Search in action with our example app in Google AI Studio. Chat with your entire image and doc library, ask questions, and trace answers back to the source:

We’re expanding the Gemini API File Search tool 🔍 with 3 new updates that enable developers to more easily build multimodal RAG systems with enhanced precision: + Multimodal Support: By leveraging our Gemini Embedding 2 model, File Search can now reason across image and text simultaneously. + Custom Metadata Filtering: Bring structure to unstructured data by tagging files with custom key-value labels. This pre-filters your data and boosts search speed. + Exact citations: File Search can now capture and return the exact source (down to the page number) for every piece of information indexed. See multimodal File Search in action with our example app in Google AI Studio. Chat with your entire image and doc library, ask questions, and trace answers back to the source:

Google AI Developers

108,801 views • 5 months ago

New short course on Building Applications with Vector Databases, taught by Pinecone’s Tim Tully! At the heart of a vector database is the ability to store a collection of vectors and then query against that, meaning input a new vector and find similar ones. This is useful for many AI applications. In this course, you'll learn how to use vector databases to build: (i) Semantic Search: Create a text search tool that goes beyond keyword matching, and instead focuses on the meaning of content. (ii) RAG (retrieval augmented generation): Enhance your LLM output by incorporating context from sources the model wasn't trained on. (iii) Recommender System: Combine semantic search and RAG to recommend topics, and demonstrate it with a news article recommender. (iv) Hybrid Search: Build an application that finds items using both images and descriptive text -- by combining both sparse and dense vector representations of the data -- using an eCommerce dataset as an example. (v) Image Similarity: Use image vector embeddings to create an app to compare facial features, using a database of public figures to determine the likeness between them. (vi) Anomaly Detection: Build an anomaly detection app that identifies unusual patterns in network communication logs. I hope you’ll enjoy learning how to build all these types of applications! Please sign up here:

New short course on Building Applications with Vector Databases, taught by Pinecone’s Tim Tully! At the heart of a vector database is the ability to store a collection of vectors and then query against that, meaning input a new vector and find similar ones. This is useful for many AI applications. In this course, you'll learn how to use vector databases to build: (i) Semantic Search: Create a text search tool that goes beyond keyword matching, and instead focuses on the meaning of content. (ii) RAG (retrieval augmented generation): Enhance your LLM output by incorporating context from sources the model wasn't trained on. (iii) Recommender System: Combine semantic search and RAG to recommend topics, and demonstrate it with a news article recommender. (iv) Hybrid Search: Build an application that finds items using both images and descriptive text -- by combining both sparse and dense vector representations of the data -- using an eCommerce dataset as an example. (v) Image Similarity: Use image vector embeddings to create an app to compare facial features, using a database of public figures to determine the likeness between them. (vi) Anomaly Detection: Build an anomaly detection app that identifies unusual patterns in network communication logs. I hope you’ll enjoy learning how to build all these types of applications! Please sign up here:

Andrew Ng

137,169 views • 2 years ago

Locally AI brings Apple’s foundation model to your iPhone, enabling on-device chat, summarization, rewriting, and more. Entirely offline and private. The model is available on Apple Intelligence-compatible devices.

Locally AI brings Apple’s foundation model to your iPhone, enabling on-device chat, summarization, rewriting, and more. Entirely offline and private. The model is available on Apple Intelligence-compatible devices.

Locally AI - Local AI Chat

42,815 views • 1 year ago

We’re introducing major updates to Gemini in @GoogleChrome for MacOS, Windows and Chromebook Plus. Built on Gemini 3, our most intelligent model, these powerful new AI features can help you multitask more easily and get the most out of the web 🧵

We’re introducing major updates to Gemini in @GoogleChrome for MacOS, Windows and Chromebook Plus. Built on Gemini 3, our most intelligent model, these powerful new AI features can help you multitask more easily and get the most out of the web 🧵

Google

2,159,782 views • 8 months ago

i can now search my entire ad library by concept. "ugc testimonials with urgency hooks" and it just finds them. here's how i built it: searchable ad swipe file > drag in creatives > AI analyzes them > search by concept > find anything in seconds -> uses google's new multimodal embedding model. images, video, and text all searchable together. -> type "ugc testimonials with urgency hooks" and it finds them. no manual tags needed. -> every ad gets auto-analyzed by gemini on import. hook type, structure, audience, CTA. all searchable. -> dual-vector scoring. visual embedding + text embedding. takes the stronger signal. full build breakdown in video + notion guide in the comments

i can now search my entire ad library by concept. "ugc testimonials with urgency hooks" and it just finds them. here's how i built it: searchable ad swipe file > drag in creatives > AI analyzes them > search by concept > find anything in seconds -> uses google's new multimodal embedding model. images, video, and text all searchable together. -> type "ugc testimonials with urgency hooks" and it finds them. no manual tags needed. -> every ad gets auto-analyzed by gemini on import. hook type, structure, audience, CTA. all searchable. -> dual-vector scoring. visual embedding + text embedding. takes the stronger signal. full build breakdown in video + notion guide in the comments

J.B.

27,430 views • 6 months ago

Meet SAM 3, a unified model that enables detection, segmentation, and tracking of objects across images and videos. SAM 3 introduces some of our most highly requested features like text and exemplar prompts to segment all objects of a target category. Learnings from SAM 3 will help power new features in Instagram Edits and Vibes, bringing advanced segmentation capabilities directly to creators. 🔗 Learn more:

Meet SAM 3, a unified model that enables detection, segmentation, and tracking of objects across images and videos. SAM 3 introduces some of our most highly requested features like text and exemplar prompts to segment all objects of a target category. Learnings from SAM 3 will help power new features in Instagram Edits and Vibes, bringing advanced segmentation capabilities directly to creators. 🔗 Learn more:

AI at Meta

194,417 views • 10 months ago

new: explore namespaces in the tpuf dashboard → browse all documents in your namespaces → check metadata and indexing status → search! text, semantic, and custom queries rolling out to all public regions over the next few days. more features coming soon

new: explore namespaces in the tpuf dashboard → browse all documents in your namespaces → check metadata and indexing status → search! text, semantic, and custom queries rolling out to all public regions over the next few days. more features coming soon

turbopuffer

31,766 views • 23 days ago