Voyage AI / voyageai/voyage-code-3

Voyage Code 3 - access through LLMTR

Voyage Code 3 is an embedding model trained for source code and technical document retrieval. It captures the relationship between function bodies, type definitions and natural-language developer queries better than general-purpose models. The 32,000-token window covers most source files in a single pass. It suits teams building code assistants, repository-wide search and technical documentation retrieval.

Technical specifications

Canonical IDvoyageai/voyage-code-3
ProviderVoyage AI
Context window32,000 tokens
OperationsEMBEDDINGS
Modalitiestext, embedding

Pricing

An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.

OperationMetricUnitPrice
EMBEDDINGSINPUT_TEXTPER_1M_TOKENS$0.180000

Example usage

With existing OpenAI SDK flows, change only the base URL and model identifier.

curl https://llmtr.com/v1/chat/completions \
  -H "Authorization: Bearer llmtr-your_key" \
  -H "Content-Type: application/json" \
  -d '{"model":"voyageai/voyage-code-3","messages":[{"role":"user","content":"Hello"}]}'

Related models