Using Vector Databases to Scale Multimodal Embeddings, Retrieval and Generation

Explore the potential of multimodal models and vector databases in scaling cross-modal retrieval and generation in this conference talk from Haystack EU 2023. Delve into the world of machine learning models that can process various data modalities, including image, video, text, audio, and tactile representations. Learn how these advanced models can address real-world problems and interact more naturally with their environment. Discover techniques for performing cross-modal retrieval and search at a billion-object scale using vector databases. Through live code demos and large-scale datasets, witness how real-time cross-modal retrieval can enhance the generative capabilities of large language models by grounding them in relevant source material. Gain insights from Zain Hasan, a Senior Developer Advocate at Weaviate, as he shares his expertise in open-source software, machine learning, and artificial intelligence applications.