Voyage: Voyage 4

Voyage 4 produces text embeddings for general-purpose and multilingual retrieval. It shares an embedding space with Voyage 4 Large and Voyage 4 Lite, making it a flexible choice for semantic search and retrieval-augmented generation.

Compare models

voyage-4

AvailableVoyageEmbeddingSync

Input / Output: $0.06 / $0.00
Context: 32K
Modalities: Embedding

About Voyage 4

Voyage 4 is a text embedding model from Voyage AI for general-purpose and multilingual retrieval. It is the middle member of the Series 4 family: its vectors are compatible with Voyage 4 Large and Voyage 4 Lite, so one index can be queried by a different family member. It returns 1024 dimensions by default, with 256, 512, and 2048 also available.

Where it works well

Vectors from Voyage 4 can be compared directly with those of Voyage 4 Large and Voyage 4 Lite.
General-purpose and multilingual retrieval quality is the stated target.
Output dimension can be reduced to 256 or 512 to shrink an index, or raised to 2048.
The input_type setting marks text as a query or a document to tune vectors for search.

When to choose another model

Voyage lists Voyage 4 Large as the model with the best retrieval quality in the series.
It embeds text only; images and code-specific retrieval need other models such as Voyage Code 3.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

EmbeddingTokenLab endpoint

POST/v1/embeddings

curl -X POST "https://api.tokenlab.sh/v1/embeddings" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "encoding_format": "float",
  "model": "voyage-4",
  "input": "The quick brown fox jumps over the lazy dog."
}'

Parameters

Embedding

Parameter	Values and limits	Default	Notes
`encoding_format`	Not provided	Not provided	This field is accepted. The current public contract does not specify a complete value range or list. No additional values are promised. The omitted-field default is not verified.
`input`	Inputs: ≥ 1, ≤ 1000	Not provided	Not provided

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Embedding

per 1M tokens

Official price: Input $0.06 / Output $0.00
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Embedding	Input $0.06 / Output $0.00	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Nov 15, 2026, 1:51 AM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Voyage 4 in Console with a prompt ready to edit or send.

Help me try voyage-4 with embedding at /v1/embeddings. Show the result and cost.

Use cases

Semantic search
Embed help articles once, then match free-form user questions against them for a search box that understands intent.
RAG retrieval
Index company documents and fetch the closest passages to give a chat model grounded context.
Multilingual lookup
Match a question written in one language to documents written in another inside the same index.

Prompt examples

Embed the query: how do I rotate an API key without downtime?

Embed this product manual section as a document for the support knowledge base.

Embed the query in German: Kündigungsfristen im Mietvertrag.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What is Voyage 4 used for?

Turning text into vectors for semantic search and retrieval-augmented generation. Voyage describes it as optimized for general-purpose and multilingual retrieval quality. You embed documents and queries, store the vectors, and rank by similarity.

Are Voyage 4 embeddings compatible with Voyage 4 Large?

Yes. Voyage states that all Series 4 embeddings are compatible with each other. That lets you index documents with one family member and embed queries with another without rebuilding the index.

Which output dimensions does Voyage 4 support?

1024 by default, with 256, 512, and 2048 as options. Smaller vectors cost less to store and search; larger ones can retain more detail. Evaluate on your own queries before choosing.

Should I set input_type?

It is recommended for retrieval. Setting it to query or document makes Voyage prepend a retrieval prompt before embedding. Voyage notes that vectors made with and without it remain compatible, so existing indexes keep working.

How much does Voyage 4 cost?

On TokenLab, Voyage 4 costs Input $0.06 / Output $0.00 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Voyage 4 use?

Use https://api.tokenlab.sh/v1/embeddings for Voyage 4. The request example below shows the matching code shape.

Which operations does Voyage 4 support?

Voyage 4 supports Embedding. Select an operation above to see its endpoint and request example.