Embedding Gemma: On-Device RAG Made Easy

In this video we learn how to use Google’s Embedding Gemma (300M) to build fast, on-device RAG with ≈200MB memory and support for 100+ languages. We will look at a RAG example. LINK: https://developers.googleblog.com/en/introducing-embeddinggemma/ https://huggingface.co/blog/embeddinggemma https://arxiv.org/pdf/2205.13147 https://huggingface.co/blog/matryoshka https://github.com/google-gemini/gemma-cookbook/blob/main/Gemma/%5BGemma_3%5DRAG_with_EmbeddingGemma.ipynb https://ai.google.dev/gemma/docs/embeddinggemma/fine-tuning-embeddinggemma-with-sentence-transformers https://ai.google.dev/gemma/docs/embeddinggemma/model_card Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-site.thinkific.com/courses/rag Let's Connect: 🦾 Discord: https://discord.com/invite/t4eYQRUcXB ☕ Buy me a Coffee: https://ko-fi.com/promptengineering |🔴 Patreon: https://www.patreon.com/PromptEngineering 💼Consulting: https://calendly.com/engineerprompt/consulting-call 📧 Business Contact: engineerprompt@gmail.com Become Member: http://tinyurl.com/y5h28s6h 💻 Pre-configured localGPT VM: https://bit.ly/localGPT (use Code: PromptEngineering for 50% off). Signup for Newsletter, localgpt: https://tally.so/r/3y9bb0 TIMESTAMPS: 00:00 EmbeddingGemma 02:15 Comparison with Other Embedding Models 02:41 Google's Interesting position 03:21 Dense Embeddings is Killing Retrieval 06:13 RAG with EmbeddingGemma 09:37 Fine-Tuning and Training the Model