Cohere Embed 5: Enterprise Frontier Embedding Models
Cohere released Embed 5 on 30 September 2026, describing it as its most powerful embedding family yet, offered in Pro and Fast tiers that share a single embedding space.
Tool Overview
Features, steps and FAQ below
Features
- ✓ Cohere says Embed 5 Pro reaches the frontier of enterprise retrieval, particularly across financial documents, parsed PDFs and visually rich documents
- ✓ Embed 5 Fast brings strong retrieval quality to latency- and cost-sensitive workloads, priced by Cohere at $0.08 per million tokens
- ✓ One index, two models: Pro and Fast share an embedding space, so teams can index with Pro and query with either model without re-indexing
- ✓ Built for complex enterprise data: Cohere says it supports multimodal inputs and retrieval, 100+ languages and a 128K-token context window
- ✓ Matryoshka representations and lower-precision outputs reduce vector storage and search costs, quantized weights lower private-deployment serving needs, and self-hosting is supported
How to Use
- Create an API key on the Cohere API or Model Vault and call embed-v5.0-pro or embed-v5.0-fast
- Index your documents using input_type=search_document
- Use input_type=search_query for queries and compute similarity to retrieve the most relevant documents
- For private deployments, use Model Vault or self-hosting, both of which support Embed 5
FAQ
What is Cohere Embed 5?
Cohere's embedding family released on 30 September 2026, which Cohere describes as its most powerful embedding models yet, offered in Pro and Fast tiers.
What is the difference between the two tiers?
Cohere says Pro is optimized for maximum retrieval quality while Fast serves latency- and cost-sensitive workloads; both share a single embedding space.
How is Embed 5 priced?
Cohere's blog notes pricing of $0.12 per million tokens for Pro and $0.08 per million tokens for Fast.
What context length and how many languages does Embed 5 support?
Cohere says it has a 128K-token context window, supports 100+ languages, and accepts multimodal inputs (text, images, and fused text + image).
Where is Embed 5 available?
Cohere says Embed 5 is generally available on the Cohere API and Model Vault, Microsoft Foundry and Amazon SageMaker.