Voyage Code 4
Voyage Code 4 is a code-specialized embedding model from Voyage AI with a 32,000 token context window.
Voyage AICode-optimized text embeddings for retrieval
Voyage Code 4 is an embedding model developed by Voyage AI, designed specifically for code-related retrieval and search tasks. It accepts text input and produces vector representations optimized for understanding and matching source code, making it well-suited for code search, semantic similarity, and retrieval-augmented generation pipelines that involve programming content. The model supports a 32,000 token context window, allowing it to embed longer code files or multi-file snippets in a single pass.
Voyage AI is a company focused on building embedding and reranking models, and Voyage Code 4 represents their code-focused offering in that lineup. The model is available as a first-party API through Voyage AI and can be accessed via MindStudio without managing separate API keys. It is best suited for developers building code search engines, documentation retrieval systems, or any application that requires dense vector representations of code.
What Voyage Code 4 supports
Code Embeddings
Generates dense vector representations tuned for source code, enabling semantic search and similarity matching across programming languages.
Long Context Support
Supports a 32,000 token context window, allowing entire code files or multi-function snippets to be embedded in a single request.
Retrieval-Augmented Generation
Produces embeddings suitable for RAG pipelines, enabling retrieval of relevant code snippets or documentation to ground downstream generation.
Semantic Similarity
Encodes code and text into a shared vector space so that semantically related queries and documents score closer together in cosine or dot-product distance.
API Integration
Available as a first-party Voyage AI API endpoint, accessible through MindStudio without requiring separate API key management.
Ready to build with Voyage Code 4?
Get Started FreeCommon questions about Voyage Code 4
What is the context window for Voyage Code 4?
Voyage Code 4 supports a context window of 32,000 tokens, which allows it to embed long code files or multi-function code blocks in a single request.
What type of model is Voyage Code 4?
Voyage Code 4 is an embedding model, meaning it converts text or code input into dense vector representations rather than generating text responses.
What is Voyage Code 4 best used for?
It is designed for code-related retrieval tasks such as semantic code search, similarity matching across programming languages, and retrieval-augmented generation pipelines that involve source code or technical documentation.
Is pricing available for Voyage Code 4?
Pricing details are not published in the current metadata. You should check the Voyage AI documentation or pricing page directly for the latest billing information.
Does Voyage Code 4 support image or video inputs?
No. Voyage Code 4 is a text-based embedding model and does not support image or video inputs.
Explore similar models
Voyage AIVoyage 4
Voyage 4 is an embedding model from Voyage AI designed for semantic search and retrieval with a 32,000-token context window.
Voyage AIVoyage 4 Large
Voyage 4 Large is an embedding model from Voyage AI with a 32,000 token context window for semantic search and retrieval.
Voyage AIVoyage 4 Lite
Voyage 4 Lite is a text embedding model from Voyage AI supporting a 32,000-token context window for semantic search and retrieval tasks.
PerplexityPerplexity Embed v1 0.6B
Perplexity Embed v1 0.6B is a quantized embedding model from Perplexity with a 32,000-token context window.
GoogleGemini Embedding 2
Gemini Embedding 2 is a text embedding model from Google designed to convert text into dense vector representations.
QwenQwen3 Embedding 0.6B
Qwen3 Embedding 0.6B is a compact embedding model from Qwen with a 32,768-token context window, served via DeepInfra.
Start building with Voyage Code 4
No API keys required. Create AI-powered workflows with Voyage Code 4 in minutes — free.