Cohere’s Rerank 3 - English is an AI model that precisely ranks documents based on their semantic similarity to a given query. It is used to enhance the performance of existing search systems. With just a few lines of code, you can augment traditional search algorithms (e.g. BM25) to boost the relevancy of results surfaced to users. Businesses also use Rerank 3 to improve the efficiency of their Retrieval-Augmented Generation (RAG) systems. It acts as a filter to ensure only the most relevant documents are being passed to the generative model. Rerank 3 is purpose-built to pair well with Cohere’s flagship generative model, Command R+.

Cohere’s Rerank 3 - English is called at inference time, using a computationally-intensive approach to outperform other search methods while adding minimal latency. It excels at ranking complex multi-aspect and semi-structured data (e.g. JSON documents, emails, code, tables).

This offer enables access to Cohere Rerank 3 - English inference API in Azure AI Studio and Azure Machine Learning. Azure AI Studio is the perfect platform for building Generative AI apps. AI Studio comes with features like Playground to explore models, Prompt Flow for prompt engineering and RAG (Retrieval Augmented Generation) to integrate your data into your apps.

The inference API is billed on a pay-as-you-go basis, per 1,000 search units.

This offer is integrated with Azure AI Studio. You will be asked to navigate to Azure AI Studio to subscribe to this offer and use the model.

Note: Parties shall count a single search unit as a query with up to 100 documents to be ranked. Documents longer than 4096 tokens when including the length of the search query will be split up into multiple chunks, where each chunk counts as a single document

Cohere Rerank 3 - EnglishCohere

Cohere Rerank 3 - English

Cohere

Cohere Rerank 3 - English

Cohere

Cohere Rerank 3 - English
Cohere