AI Services Directory

Embedding / RAG · Updated 2026-10-25

Voyage AI — what it is actually for

Voyage AI, part of MongoDB, provides text, multimodal, code, finance, and law embedding models plus rerankers. Usage is billed per token after a large one-time free allowance.

Short answer: use Voyage AI when you need rag retrieval with domain-specific embeddings. It is a free token allowance + usage-based api service from Voyage AI (MongoDB); the official pricing source is linked below.

Vendor
Voyage AI (MongoDB)
Category
Embedding / RAG
Pricing model
Free token allowance + usage-based API
Free option
Yes. Each account gets a one-time free allowance, such as 200M tokens for current models.

What Voyage AI is used for

  • RAG retrieval with domain-specific embeddings
  • Code search and technical documentation retrieval
  • Reranking search results

Source: official product page and official documentation.

Who Voyage AI is for

  • Teams that need domain-specific embeddings
  • MongoDB-based RAG architectures
  • Cost-sensitive retrieval at scale

Source: official documentation.

Who it is not for

  • Buyers who need a monthly free reset
  • Consumer use without development resources

Source: documented plans, limits, and deployment model on the official product page.

Pricing

  • The first 200M tokens are free for current text embedding models such as voyage-4-large, voyage-4, voyage-4-lite, voyage-context-4, and voyage-code-3.Source: docs.voyageai.com
  • After the free allowance, text embeddings cost US$0.02 to US$0.12 per 1M tokens depending on the model.Source: docs.voyageai.com
  • Multimodal embeddings are billed per text token and per pixel; rerankers cost US$0.02 to US$0.05 per 1M tokens. Batch API usage does not use free credits.Source: docs.voyageai.com

The free allowance is described as a first-N-tokens allowance, not a monthly reset. Older models have smaller or no free allowances.

Full pricing page: https://docs.voyageai.com/docs/pricing

Alternatives to Voyage AI

  • OpenAI Embeddings — Text embedding models billed per input token for search and RAG.
  • Cohere Embed — Embed 5 Pro and Fast models with separate indexing and query pricing.

Official sources

Voyage AI FAQ

Is Voyage AI free?

Each account receives a one-time free token allowance. For current models the allowance is 200M tokens; older specialized models have smaller allowances.

Is the free allowance monthly?

No. Voyage describes the allowance as the first N tokens free for every account, not a monthly reset.

Does Voyage have rerankers?

Yes. Voyage publishes rerank models priced per token, with a separate free allowance.

See also: compare all services by use case · about this directory.