Skip to main content
Voyage AI (MongoDB)

Voyage AI (MongoDB): Foundation Models

Founded 2023|Palo Alto, CA|11-50 people|Usage-based

Respan Gateway: up to 55% off tokens on Gemini, GPT, Claude and 1,000+ more models. One API key, 0% fee.

What is Voyage AI (MongoDB)?

Key features

Core capabilities this platform advertises.

  • Text & multimodal embeddings
  • Domain-specific models (code, law, finance)
  • Rerankers for retrieval quality
  • MoE architecture embeddings
  • Open-weight nano model

Strengths and tradeoffs

What this tool does well, and the limitations to keep in mind.

Pros

  • Best-in-class retrieval accuracy — consistently outperforms OpenAI and Cohere by 10-20% across 100+ benchmark datasets
  • Generous free tier with 200M free tokens — enough to embed a sizable corpus before paying
  • Domain-specific models for code, law, and finance that outperform general-purpose alternatives by ~15%
  • Industry-leading 32K context window — 4x OpenAI's limit, enabling full-document embeddings without chunking
  • Shared embedding space across Voyage 4 family — use cheaper models for queries, expensive ones for indexing

Cons

  • API-only with no self-hosting option except for the open-weight voyage-4-nano model
  • MongoDB acquisition creates vendor lock-in risk as roadmap priorities may increasingly favor Atlas-native workflows
  • API round-trip adds latency compared to local embedding models
  • Specialized model pricing at $0.18/M tokens is significantly more expensive than OpenAI alternatives

Plans & pricing

What's included in each plan, and how the tiers compare.

Free Tier

200M free tokens

 

  • Access to all current models
  • Embedding and reranking

voyage-4-large

$0.12/M tokens

Usage-based

  • Best accuracy
  • MoE architecture
  • Up to 2048 dimensions

voyage-4

$0.06/M tokens

Usage-based

  • Mid-tier performance
  • Shared embedding space

voyage-4-lite

$0.02/M tokens

Usage-based

  • Cost-efficient
  • Good for high-volume

Rerankers

$0.02-0.05/M tokens

Usage-based

  • rerank-2.5 and rerank-2.5-lite
View official pricing page

Common use cases

Developers building search and RAG applications

  • Semantic search
  • RAG retrieval
  • Code search
  • Legal document search
  • Multimodal retrieval

Using Voyage AI (MongoDB) with Respan

Voyage AI embeddings power the retrieval layer in RAG applications, while Respan monitors the LLM calls that consume those retrieved results. Together, they provide end-to-end observability from document retrieval through final LLM response generation.

  • Monitor RAG pipeline performance from Voyage AI embedding to LLM response via Respan tracing
  • Track embedding API costs alongside LLM inference costs in a unified Respan dashboard
  • Evaluate retrieval quality by correlating Voyage AI search results with LLM output quality in Respan
  • Optimize the full RAG stack by identifying bottlenecks across embedding, retrieval, and generation stages

Run Voyage AI (MongoDB) in production with Respan

Up to 55%Off tokens
1,000+Models
0%Fee

Respan lets you trace LLM and agent calls across any model or framework, A/B test prompts on production traffic, and route requests across 1,000+ models through one gateway.

Best Voyage AI (MongoDB) alternatives & competitors

Top companies in Foundation Models you can use instead of Voyage AI (MongoDB).

View all Voyage AI (MongoDB) alternatives

Compare Voyage AI (MongoDB)

Side-by-side comparisons with other tools in this category.

Best integrations for Voyage AI (MongoDB)

Companies from adjacent layers in the AI stack that work well with Voyage AI (MongoDB).

Last verified: March 27, 2026