Cohere · Embedding
Embed v4
Cohere Embed v4 is a state-of-the-art embedding model supporting 100+ languages with best-in-class retrieval performance.
Evidence and sources
Maker HQ: US (according to Cohere's own website)
Sources: Cohere API data, AWS Bedrock Runtime data, Heroku Managed Inference data, Models.dev data, EURouter dataPrimary data from Cohere API
from AWS Bedrock Runtime: name
from Heroku Managed Inference: primary mode
from Models.dev: weights availability
from EURouter: description
Tags from its sources:proprietaryembeddingmultilingualretrieval
Advantages and trade-offs
Compared with 10 representative embedding models
Advantages
Trade-offs
Costs more than most peers
No open weights
The weights aren't public, so you can't run it yourself.
Input & output
as stated by the model's providersText
Images
Embeddings
Also known as: azure_ai/embed-v-4-0, cohere-embed-v-4-0, cohere-embed-v4, cohere.embed-v4:0, cohere/embed-v4.0, embed-v-4-0, embed-v4.0, eu.cohere.embed-v4:0, global.cohere.embed-v4:0, oci/cohere.embed-v4.0, us.cohere.embed-v4:0, vercel_ai_gateway/cohere/embed-v4.0
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works