{
  "slug": "voyage-ai",
  "name": "Voyage AI",
  "aliases": ["Voyage AI by MongoDB", "voyageai"],
  "vendor": { "name": "MongoDB (Voyage AI)", "slug": "mongodb", "website": "https://www.voyageai.com" },
  "homepage_url": "https://www.voyageai.com/",
  "docs_url": "https://docs.voyageai.com/docs/introduction",
  "repo_url": "https://github.com/voyage-ai/voyageai-python",
  "pricing_url": "https://docs.voyageai.com/docs/pricing",
  "product_type": "api-service",
  "product_status": "active",
  "summary": "Hosted embedding and reranking models for retrieval, including code, finance, law and multimodal variants, offered by MongoDB with a free token allowance and pay-as-you-go API",
  "description": "Voyage AI provides embedding and reranking models as a hosted API and is now branded as part of MongoDB's data platform. The embedding endpoint converts text, code, images and video into dense vectors for semantic search and retrieval-augmented generation; the rerank endpoint scores query-document pairs to refine candidates from vector or lexical search. Current families include the voyage-4 text series in large, standard and lite sizes plus a context-focused variant, voyage-code for source code, voyage-finance and voyage-law for domain text, voyage-multimodal for mixed media and the rerank series. A Batch API processes large jobs at lower cost and a Files API stores inputs.\n\nAccess is through a REST API with an official Python client (MIT) and through cloud marketplaces such as AWS, including in-VPC deployment; the models are designed to work with any vector database, not only MongoDB Atlas. Pricing is a free token allowance per account followed by per-token (and per-pixel for images) billing, with fine-tuned models on dedicated instances arranged through sales. Voyage is a natural pairing for a retrieval stack that wants stronger domain embeddings than general-purpose providers, with the usual trade-off that embeddings are computed off-premises and tied to a vendor's model lifecycle.",
  "capabilities": ["Text embedding models in several sizes including a context-focused variant", "Domain embeddings for code, finance and law", "Multimodal embeddings for text, images and video", "Rerankers for query-document relevance scoring", "Batch API and Files API for large jobs", "Python client and marketplace or in-VPC deployment"],
  "workflows": ["Embed a document corpus and store vectors in a database", "Rerank the top results of a hybrid search before passing them to an LLM", "Run a nightly batch embedding job over new filings"],
  "advantages": ["Domain-specific and multimodal model lineup", "Free allowance for prototyping", "Vector-database agnostic"],
  "limitations": ["Hosted only; data leaves your infrastructure", "Model deprecations require re-embedding corpora", "Fine-tuned and dedicated options go through sales"],
  "prerequisites": ["Voyage AI API key", "A vector store for the resulting embeddings"],
  "audiences": ["Teams building RAG over specialized corpora", "Developers on MongoDB Atlas", "Researchers needing finance or legal embeddings"],
  "alternatives": ["cohere", "openai-platform", "sentence-transformers", "gemini-api"],
  "categories": [{ "slug": "embeddings-vector-search", "primary": true }],
  "tags": ["embeddings", "vector-search", "rag", "api", "reranking", "python"],
  "use_cases": ["build-agents"],
  "operating_systems": ["any"],
  "execution_modes": ["cloud"],
  "interfaces": ["api", "sdk"],
  "deployment_models": ["saas", "managed-cloud"],
  "integrations": ["MongoDB Atlas", "AWS Marketplace", "LangChain", "LlamaIndex"],
  "data_formats": ["JSON", "Text", "Images", "Video"],
  "languages": ["Python"],
  "source_model": "proprietary",
  "pricing_model": ["freemium", "usage-based", "enterprise"],
  "pricing_checked_at": "2026-11-02",
  "self_hostable": false,
  "sources": [
    { "url": "https://www.voyageai.com/", "type": "official_site", "title": "Voyage AI by MongoDB", "supports": ["name", "vendor", "summary"], "checked_at": "2026-11-02" },
    { "url": "https://docs.voyageai.com/docs/introduction", "type": "official_docs", "title": "Voyage AI documentation", "supports": ["capabilities"], "checked_at": "2026-11-02" },
    { "url": "https://docs.voyageai.com/docs/pricing", "type": "pricing_page", "supports": ["pricing_model", "capabilities"], "checked_at": "2026-11-02", "note": "Free allowance then pay-as-you-go; lists voyage-4, code, finance, law, multimodal and rerank families" },
    { "url": "https://github.com/voyage-ai/voyageai-python", "type": "repository", "title": "voyageai-python (MIT)", "supports": ["languages"], "checked_at": "2026-11-02" }
  ],
  "confidence": "high",
  "discovery_method": "editorial-research",
  "depth": "standard",
  "editorial_status": "reviewed",
  "visibility": "public",
  "featured": false,
  "author": "batch-A"
}