Cohere Embed 4
Cohere Embed 4 is an embedding model from Cohere with a 128,000-token context window, released in April 2025.
CohereMultimodal embeddings with a 128K context window
Cohere Embed 4 (embed-v4.0) is an embedding model developed by Cohere and released in April 2025. It converts text into dense vector representations that can be used for semantic search, retrieval-augmented generation, classification, and clustering tasks. With a 128,000-token context window, it can process long documents in a single pass without chunking strategies that might otherwise lose context.
Embed 4 is designed for enterprise retrieval use cases where accurate semantic similarity across large corpora is important. It supports multimodal inputs, including both text and images, making it applicable to pipelines that need to index and retrieve mixed-content documents. Developers typically use it as the retrieval backbone in RAG systems, recommendation engines, and document search applications.
What Cohere Embed 4 supports
Long Context Embedding
Encodes documents up to 128,000 tokens into a single embedding vector, reducing the need for document chunking in retrieval pipelines.
Semantic Search
Generates dense vector representations that enable similarity-based retrieval across large text corpora using nearest-neighbor search.
Multimodal Input Support
Accepts both text and image inputs, allowing mixed-content documents to be embedded and retrieved within a unified vector space.
RAG Retrieval
Serves as the embedding backbone for retrieval-augmented generation pipelines, indexing documents so a language model can retrieve relevant context at inference time.
Classification & Clustering
Produces embeddings suitable for downstream classification and clustering tasks by capturing semantic relationships between inputs in vector space.
Ready to build with Cohere Embed 4?
Get Started FreeCommon questions about Cohere Embed 4
What is the context window for Cohere Embed 4?
Cohere Embed 4 supports a context window of 128,000 tokens, allowing long documents to be embedded in a single call.
What types of inputs does Cohere Embed 4 accept?
Cohere Embed 4 supports both text and image inputs, making it suitable for multimodal retrieval pipelines.
What is Cohere Embed 4 best used for?
It is designed for semantic search, retrieval-augmented generation (RAG), document classification, and clustering tasks where accurate vector representations of content are needed.
What is the pricing for Cohere Embed 4?
Pricing information is not included in the available metadata. Refer to Cohere's official pricing page at cohere.com/pricing for current rates.
When was Cohere Embed 4 released?
Cohere Embed 4 was released in April 2025.
Explore similar models
CohereCohere Rerank 4 Fast
Cohere Rerank 4 Fast is a reranking model from Cohere designed to score and reorder search results with a 4096 token context window.
CohereCohere Rerank 4 Pro
Cohere Rerank 4 Pro is a reranking model from Cohere designed to improve search result relevance with a 4096-token context window.
CohereCohere Rerank 3.5
Cohere Rerank 3.5 is a reranking model from Cohere that reorders search results by semantic relevance with a 4096 token context window.
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.
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.
Start building with Cohere Embed 4
No API keys required. Create AI-powered workflows with Cohere Embed 4 in minutes — free.