Voyage AI

aid: voyage-ai
url: https://raw.githubusercontent.com/api-evangelist/voyage-ai/refs/heads/main/apis.yml
name: Voyage AI
kind: company
description: >-
  Voyage AI builds state-of-the-art embedding and reranker models for
  retrieval-augmented generation (RAG) and semantic search. The platform
  exposes an OpenAI-style REST API at api.voyageai.com/v1 for text embeddings,
  multimodal embeddings, contextualized embeddings, and reranking, with
  Python and TypeScript SDKs. Model families include voyage-3.x and voyage-4.x
  text embeddings, voyage-code-3, domain-specialised models (voyage-finance-2,
  voyage-law-2), voyage-multimodal-3, and the voyage-rerank-2 reranker family.
  Voyage AI was acquired by MongoDB in February 2024 and is integrated into
  MongoDB Atlas Vector Search; models are also distributed via AWS Marketplace,
  Azure Marketplace, and Snowflake.
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
  - Embeddings
  - Rerankers
  - RAG
  - Semantic Search
  - AI Models
  - Vector Search
  - Multimodal
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
apis:
  - aid: voyage-ai:embeddings
    name: Voyage AI Embeddings API
    description: >-
      OpenAI-compatible REST endpoint that returns dense vector embeddings for
      input text. Supports model selection (voyage-3.5, voyage-3-large,
      voyage-code-3, voyage-finance-2, voyage-law-2, voyage-4 family),
      configurable output dimensions (256, 512, 1024, 2048), output dtype
      (float, int8, uint8, binary, ubinary), input_type hints (query or
      document), and batch sizes up to 1000 inputs per request.
    humanURL: https://docs.voyageai.com/reference/embeddings-api
    baseURL: https://api.voyageai.com/v1
    tags:
      - Embeddings
      - Text
      - REST
    properties:
      - type: Documentation
        url: https://docs.voyageai.com/reference/embeddings-api
      - type: OpenAPI
        url: https://docs.voyageai.com/llms.txt
  - aid: voyage-ai:rerank
    name: Voyage AI Rerank API
    description: >-
      Reranking endpoint that scores a list of candidate documents against a
      query and returns relevance scores. Powered by the voyage-rerank-2 model
      family, used downstream of vector search to improve retrieval precision
      in RAG pipelines.
    humanURL: https://docs.voyageai.com/reference/reranker-api
    baseURL: https://api.voyageai.com/v1
    tags:
      - Rerank
      - Retrieval
      - RAG
    properties:
      - type: Documentation
        url: https://docs.voyageai.com/reference/reranker-api
  - aid: voyage-ai:multimodal-embeddings
    name: Voyage AI Multimodal Embeddings API
    description: >-
      Multimodal embeddings endpoint backed by voyage-multimodal-3 that accepts
      interleaved text and images in a single request and returns embeddings in
      a shared vector space, enabling cross-modal retrieval for documents that
      mix text, screenshots, charts, and figures.
    humanURL: https://docs.voyageai.com/reference/multimodal-embeddings-api
    baseURL: https://api.voyageai.com/v1
    tags:
      - Embeddings
      - Multimodal
      - Vision
    properties:
      - type: Documentation
        url: https://docs.voyageai.com/reference/multimodal-embeddings-api
  - aid: voyage-ai:contextualized-embeddings
    name: Voyage AI Contextualized Embeddings API
    description: >-
      Endpoint that embeds chunks while conditioning on surrounding document
      context, improving recall for long-document RAG workflows where chunk
      embeddings would otherwise lose document-level signal.
    humanURL: https://docs.voyageai.com/reference/contextualized-embeddings-api
    baseURL: https://api.voyageai.com/v1
    tags:
      - Embeddings
      - Contextualized
      - RAG
    properties:
      - type: Documentation
        url: https://docs.voyageai.com/reference/contextualized-embeddings-api
  - aid: voyage-ai:python-sdk
    name: Voyage AI Python SDK
    description: >-
      Official Python client (voyageai) wrapping the embeddings, multimodal,
      contextualized, and reranking endpoints with batching, retries, and
      async support.
    humanURL: https://github.com/voyage-ai/voyageai-python
    baseURL: https://github.com/voyage-ai/voyageai-python
    tags:
      - SDK
      - Python
    properties:
      - type: Repository
        url: https://github.com/voyage-ai/voyageai-python
      - type: Package
        url: https://pypi.org/project/voyageai/
  - aid: voyage-ai:typescript-sdk
    name: Voyage AI TypeScript SDK
    description: >-
      Official TypeScript / JavaScript client for the Voyage AI REST API.
    humanURL: https://github.com/voyage-ai/typescript-sdk
    baseURL: https://github.com/voyage-ai/typescript-sdk
    tags:
      - SDK
      - TypeScript
      - JavaScript
    properties:
      - type: Repository
        url: https://github.com/voyage-ai/typescript-sdk
common:
  - type: LinkedIn
    url: https://www.linkedin.com/company/voyage-ai
  - type: Website
    url: https://www.voyageai.com/
  - type: Documentation
    url: https://docs.voyageai.com/
  - type: GitHub
    url: https://github.com/voyage-ai
  - type: Pricing
    url: https://docs.voyageai.com/docs/pricing
  - type: Parent
    url: https://www.mongodb.com/
  - type: Plans
    url: plans/voyage-ai-plans-pricing.yml
  - type: RateLimits
    url: rate-limits/voyage-ai-rate-limits.yml
  - type: FinOps
    url: finops/voyage-ai-finops.yml
integrations:
  - name: MongoDB Atlas Vector Search
  - name: AWS Marketplace
  - name: Azure Marketplace
  - name: Snowflake
  - name: LangChain
  - name: LlamaIndex
  - name: Haystack
maintainers:
  - FN: Kin Lane
    email: [email protected]