TwelveLabs

We're transforming film production workflows with cutting-edge AI. ?? Forget the manual grind of sorting daily rushes; our Marengo model automatically organizes your footage. Imagine quickly categorizing a "Game of Thrones" style epic into scenes like “Battle Sequences” or “Dialogue in the Throne Room” at the click of a button! ?? What’s more? Our Pegasus model sifts through these scenes to highlight key moments, create summaries, and even suggests captions or titles, speeding up the process of repurposing or reviewing content dramatically. ?? Dive deeper into how our technology not only supports but amplifies the creative process, ensuring that artistry always takes center stage. Check out our blog for more on how Twelve Labs is setting new standards in film production. Read all about it here:?https://lnkd.in/ghndYHWF #VideoAI #TwelveLabs

Ready to level up your video AI game? Join TwelveLabs and Pinecone for an insightful webinar on "Mastering Video AI: Contextual Advertising and Personalized Ads Recommendation." ?? Catch James Le from TwelveLabs and Arjun Patel from Pinecone as they unpack the magic behind video foundation models like Marengo and Pegasus. They'll show you how these models transform advertising and content personalization. Here’s what you’ll dive into: ? The nuts and bolts of innovative video foundation models. ? Integrating video embeddings effectively with Pinecone's vector database. ? A live demo on creating contextual ads and discovering personalized content. Secure your spot now, and even if you can't join live on April 7, register to get the recording! ?? Sign up here:?https://lnkd.in/g6qJW8VQ #TwelveLabs #VideoAI

We are excited to announce our technical integration with Snowflake, enabling developers to unlock powerful video understanding capabilities at scale! At TwelveLabs, we are committed to making video as searchable and analyzable as text. The integration with Snowflake Cortex demonstrates this commitment by bringing our state-of-the-art video embedding model directly into Snowflake's AI Data Cloud. ?? We have published a detailed technical guide showing developers how to: ?? - Harness our multimodal embeddings within Snowflake's infrastructure - Implement efficient video processing using Snowpark Container Services - Build sophisticated video search applications with our Marengo 2.7 model - Create interactive Streamlit applications combining our API with Snowflake Cortex What sets this integration apart is how it seamlessly combines our video-native models with Snowflake's enterprise-grade capabilities. Developers can now leverage our advanced embedding technology while maintaining the security and scalability that Snowflake provides. ??? The possibilities are wide-ranging – from contextual ad placement and content moderation to sophisticated video search engines and recommendation systems. All are powered by TwelveLabs' cutting-edge video understanding technology, which is now available via Snowflake interface. ?? Ready to upgrade your video applications? Start building with TwelveLabs and Snowflake today! Check out our complete tutorial in the comments ??

~ New Webinar ~ Check out #MultimodalWeekly 76 recording with Simon Lecointe, Jeremy Care, and Evan Elezaj: https://lnkd.in/gz32_nDS ?? They discussed how Generative AI is transforming highlight reel production for sports and entertainment - showcasing a live demo using technologies from TwelveLabs, TrackIt, and Amazon Web Services (AWS). ?? ? ??? Enjoy!

What an incredible week at #GTC2025! We're still buzzing from the energy and excitement. A massive thank you to everyone who visited our booth —you truly made our experience memorable. ?? A special shout-out to our CEO, Jae Lee, whose inspiring panel captivated attendees and sparked meaningful conversations about the future of AI in video technology. Thank you to the NVIDIA GTC team and all the attendees for making this event a success. We're already looking forward to next year! #TwelveLabs #VideoAI #GTC

In the 77th?session of #MultimodalWeekly, we have three exciting presentations on video frame interpolation, video restoration, and multi-shot video understanding. ??Zujin GUO?will present?Generalizable Implicit Motion Modeling?(GIMM), a novel and effective approach to motion modeling for Video Frame Interpolation: https://lnkd.in/eaCFV3Uc ??Kamran Janjua?and?Amirhosein Ghasemabadi?will present?Turtle, a method to learn the truncated causal history model for efficient and high-performing video restoration: https://lnkd.in/gBukU-Aj ??Mingfei Han?will present?Shot2Story, a new multi-shot video understanding benchmark dataset with detailed shot-level captions, comprehensive video summaries, and question-answering pairs: https://lnkd.in/gXTSg4QW Register for the webinar here: https://lnkd.in/gJGtscSH ?? Join our Discord community: discord.gg/mwHQKFv7En ??

?? Excited to share our latest technical deep-dive on building a Multimodal RAG system using TwelveLabs and Chroma! We have created a comprehensive guide that walks through implementing video-based RAG, perfect for developers looking to enhance their applications with powerful video understanding capabilities. ??? Technical Implementation Highlights: ? Set up TwelveLabs API for video processing and understanding ? Configure Chroma vector database for efficient video segment storage ? Implement semantic search across video content ? Compare performance between TwelveLabs Pegasus and open-source LLaVA-NeXT-Video models ?? Key Benefits of TwelveLabs + Chroma Integration: ? Simplified video indexing and retrieval ? Rich understanding of video content without complex infrastructure ? Efficient local vector storage with Chroma ? Seamless integration with existing RAG pipelines ?? What makes this integration powerful: ?? Process and index entire video libraries ?? Extract meaningful insights from video segments ?? Build conversational interfaces for video content ?? Scale with minimal operational overhead Check out the complete tutorial with code samples and implementation details in the comment below. Perfect for AI engineers and developers working on multimodal applications!

?? NAB Show 2025 is just around the corner! Join TwelveLabs in Las Vegas, April 5-9, for a deep dive into the latest innovations reshaping the media and entertainment industry. ?? From main stage keynotes to cutting-edge demos in the Creator Lab, this year's National Association of Broadcasters is packed with innovation. We’ll be announcing our panel sessions soon—keep an eye out for updates! ?? Stop by our booth W3921 to get a glimpse into the future of video intelligence that's transforming how content creators work, analyze, and monetize their media. ?? Want to connect? Schedule time with our team now and be first in line for personalized demos: https://lnkd.in/gTd9cwxH #NABShow2025 #TwelveLabs #VideoAI

Just back from HumanX! The energy was electric as our co-founder Soyoung Lee took the stage to discuss "Aligning Human Expertise with AI Infrastructure." Soyoung unpacked the real magic of TwelveLabs’ AI—transforming how we handle video content. Imagine shifting from manually logging hours of footage to letting AI do the heavy lifting. This doesn't just speed things up; it revolutionizes content creation, allowing teams to focus on crafting more engaging, customized stories. Soyoung also broke down how to smartly integrate AI: start simple, build on what works, and always aim to enhance what your team can do with AI’s help. She also touched on the practical steps for adopting AI, like the importance of cloud migration to leverage new technologies fully and how foundational models can set the stage for more specific, impactful applications. Thanks to everyone who came out to listen and engage. Check out some highlights from the event! ?? #TwelveLabs #VideoAI #HumanX2025

We are thrilled to share our latest collaboration with Weaviate, demonstrating how TwelveLabs' advanced video understanding capabilities can revolutionize video processing through Retrieval-Augmented Generation (RAG). We have put together a comprehensive technical guide that showcases the power of our combined technologies. ?? At the heart of this integration is our Pegasus model for sophisticated video question answering, working in concert with our Marengo model for generating precise multimodal video embeddings. When paired with Weaviate's vector search capabilities, we've created a solution that transforms how developers can work with video content. Through extensive testing, including experiments with LLaVa-NeXT-Video for multi-video RAG, we've demonstrated the robust capabilities of our combined approach. ?? The performance improvements speak for themselves. Our integration delivers significant reductions in processing times while maintaining high accuracy in video content understanding. What's particularly exciting for us at TwelveLabs is seeing how this architecture scales effectively for real-world applications, making it a valuable solution for enterprise-level implementations. ??? We invite developers to explore the full technical article, complete with code samples and performance metrics, to see how TwelveLabs' video AI capabilities, combined with Weaviate's vector database, can transform your video processing workflows. ??