Amazon Bedrock now offers multimodal support for Cohere Embed 3 Multilingual and Embed 3 English
Amazon Bedrock now offers multimodal support for Cohere Embed 3 Multilingual and Embed 3 English, foundation models that generate embeddings from both text and images. This powerful addition to Amazon Bedrock can enable enterprises to unlock significant value from their vast amounts of data, including visual content. With this new capability, businesses can build systems that accurately and quickly search important multimodal assets such as complex reports, product catalogs, and design files.
According to Cohere, Embed 3 delivers exceptional performance on various retrieval tasks and is engineered to handle diverse data types. Supporting search functionality for both text and images, and in over 100 languages (Embed 3 Multilingual), it is well-suited for global applications. These models are designed to process and interpret varied datasets, effectively managing inconsistencies typical in real-world scenarios. This versatility makes Embed 3 particularly valuable for enterprises seeking to enhance their search and retrieval systems across different data formats. By leveraging this technology, businesses can develop more comprehensive search applications, and can lead to improved user experiences and increased efficiency across various use cases.
Cohere Embed 3 with multimodal support is now available in Amazon Bedrock and is supported in 12 AWS Regions, for more information on supported Regions, visit the Amazon Bedrock Model Support by Regions guide. For more details about Cohere Embed 3 and its capabilities, visit the Cohere product page. To get started with Cohere Embed 3 in Amazon Bedrock, visit the Amazon Bedrock console.