Skip to main content
Modelembeddingsv4.0

Cohere Embed V4

by Cohere · paid · Last verified 2026-03-17

Cohere Embed V4 is Cohere's fourth-generation embedding model supporting both text and image inputs for unified multimodal retrieval. It delivers state-of-the-art performance on MTEB benchmarks and is optimized for enterprise RAG pipelines, semantic search, and cross-modal retrieval applications.

https://cohere.com/embed
B
BAbove Average
Adoption: BQuality: A+Freshness: A+Citations: BEngagement: F

Specifications

License
Proprietary
Pricing
paid
Capabilities
text-embeddings, image-embeddings, semantic-search, retrieval, cross-modal-search
Integrations
Cohere API, Weaviate, Pinecone, AWS Bedrock, Azure AI
Use Cases
rag-pipelines, semantic-search, document-retrieval, multimodal-search, recommendation-systems
API Available
Yes
Parameters
Unknown
Context Window
128K
Modalities
text, image
Training Cutoff
2025
Tags
cohere, embeddings, multimodal, enterprise, retrieval
Added
2026-03-17
Completeness
100%

Index Score

60.7
Adoption
68
Quality
90
Freshness
96
Citations
62
Engagement
0

Explore the full AI ecosystem on Agents as a Service