twelvelabs

Official TwelveLabs SDK for Python

Official TwelveLabs SDK for Javascript

AI agent that automates any repetitive browser task from screen recordings

How TwelveLabs AI Models on Amazon Bedrock and OpenSearch enable multi-vector semantic search for video content.

AI-powered video analysis pipeline using Amazon Bedrock TwelveLabs Pegasus 1.2 and Claude 3.7 Sonnet for automated construction site monitoring and educational content categorization.

Demonstrating the use of Amazon Nova Multimodal Embeddings and TwelveLabs Pegasus 1.2 models on Amazon Bedrock along with Amazon OpenSearch Serverless to perform semantic search.

AI-powered real-estate platform with a focus on video analysis & interaction.

AI-powered 3D video exploration platform with multilingual narration in your own voice.

Multimodal AI embeddings with TwelveLabs Marengo on Amazon Bedrock. Generate unified vector representations for text, video, audio, and images. Features cross-modal search, S3 Vector Store integration, and production-ready patterns for semantic similarity search.

Explainable AI decision layer for less disruptive long-form video advertising.

agentic ai for shopify merchants to protect/express/market their brand identity (winner at uofthacks 2026)

Answers what's in this video that you don't already know — your notes and the video segments in one Neo4j graph, so 'what should I watch?' is a set difference that returns a timecoded cut list.

Real-time AI agent for querying live courtroom video with sub-500ms latency. Multimodal search combining video intelligence, speech-to-text, and hybrid search. Built with Stream, Twelve Labs, Deepgram, and Gemini Live API.

Does resolution matter to a video foundation model ? A 5-experiment, 2,700 query study of TwelveLabs Marengo accross 360p-4K.

app that helps you detect various events in the backend video.

Turn multi-source body-cam, dashcam and document evidence into a temporal Neo4j knowledge graph, then ask investigative questions across it and get answers where every statement cites a playable video timecode. Strands + TwelveLabs + OpenAI.

Natural-language search inside videos via the TwelveLabs API — Streamlit UI, FAISS hybrid retrieval, and an MRR/P@K evaluation harness

Video context graph: keeps what's said, shown and written on screen apart, then finds where channels contradict each other. TwelveLabs + OpenAI + Neo4j + Strands.

Natural language video search on TwelveLabs Marengo 2.7. Searches visual, audio, and conversation tracks, returns ranked clips with timestamps and transcripts.