Marqo Open Source

Marqo Open Source is a self-hosted search engine for developers building document search, product discovery, or retrieval for AI applications. The project is deprecated and no longer receives updates.

• Text and image search: Search documents by meaning or keywords, and use Hugging Face CLIP models to find images with text or image queries.
• Built-in embeddings: The engine handles embedding generation, document and metadata storage, and retrieval through one API, so you don't have to supply precomputed vectors.
• Search controls: Combine weighted query terms, apply negative weights to unwanted concepts, and filter results using document metadata.
• Model choice: Use a preconfigured model or bring your own, with support for PyTorch, Hugging Face, and OpenAI models.
• AI framework integrations: Connect retrieval to Haystack, Griptape, LangChain, or Hamilton for question answering and other applications that search your own data.

The open-source engine runs in Docker, supports CPU and GPU processing, and uses the Apache 2.0 license. It requires at least 8 GB of memory and 50 GB of storage allocated to Docker. Image indexing accepts files from local disk or internet URLs.

Marqo's current ecommerce search and discovery platform is a separate product from this discontinued open-source engine.