Cohere Embed 5 and Rerank 5 arrive on OpenRouter’s unified API
OpenRouter users can call Cohere’s Embed 5 and Rerank 5 models with their existing API key; four new endpoints cover text, image and mixed-content RAG workflows.
AT A GLANCE
Key points
- Four Embed 5 and Rerank 5 endpoints are available on OpenRouter.
- Embed 5 supports text, image and mixed inputs.
- The Pro and Fast embedding models share one vector space.
What is changing?
On November 4, Cohere made the Embed 5 and Rerank 5 families available through OpenRouter. Users can call four models with their existing OpenRouter API key and credits without setting up a separate Cohere account or SDK.
Embed 5 Pro targets quality-critical indexing, while Embed 5 Fast is positioned for low-latency queries and high-volume agent loops. The two versions share an embedding space, allowing teams to index a corpus with Pro and embed queries with Fast. Embed 5 accepts text, images and mixed inputs such as PDF pages.
Rerank 5 Pro targets complex queries and long documents, while Rerank 5 Fast focuses on latency- and cost-sensitive ranking. Provider performance claims should be validated independently on production datasets.
Explore the original sources
Source date: 4 November 2026.
Published by AI Radar: 4 Nov 2026, 22:02 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…