Embeddings (Vectors, Search, Retrieval)
Introduction to Embeddings at Cohere
Embeddings transform text into numerical data, enabling language-agnostic similarity searches and efficient storage with compression.
Semantic Search with Embeddings
Examples on how to use the Embed endpoint to perform semantic search (API v2).
Unlocking the Power of Multimodal Embeddings
Multimodal embeddings convert text and images into embeddings for search and classification (API v2).
Batch Embedding Jobs with the Embed API
Learn how to use the Embed Jobs API to handle large text data efficiently with a focus on creating datasets and running embed jobs.
Using the Embed API
This document provides a guide on using the Cohere Embed API endpoint to generate embeddings for text data, with instructions on setting up the SDK, running a sample Question and Answering scenario, and details on input types and compression levels for the embeddings.
Batch Embedding Jobs with the Embed API
Learn how to use the Embed Jobs API to handle large text data efficiently with a focus on creating datasets and running embed jobs.
Bulk Embedding (might be redundant)
This document explains how to efficiently embed text in bulk using an endpoint that returns a JSONL file of text embeddings with their respective text, which can be downloaded within 2 months.
Embedding Large Datasets
Multilingual Embed Models
Cohere offers multilingual language models that map text to a semantic vector space, improving search results and enabling use cases such as multilingual semantic search, customer feedback aggregation, and cross-lingual content moderation. The model outperforms other models in clustering, search, and cross-lingual classification tasks.
Cross-Lingual Content Moderation
The document discusses the challenge of content moderation in a multilingual online environment and proposes using multilingual embeddings to create a tool that can work across 100+ languages with training data in English.
Customer Feedback Aggregation
This document discusses how successful products like the iPhone receive thousands of reviews in multiple languages, and how using Cohere's multilingual model can help companies analyze and understand customer feedback across different languages and markets.
Multilingual Semantic Search
Semantic search now works across languages, allowing for the retrieval of relevant information regardless of the language in which it was published, such as in the financial domain.
Supported Languages
A list of languages that Cohere's multilingual embedding model provides.
Unlocking the Power of Multimodal Embeddings
Multimodal embeddings convert text and images into embeddings for search and classification.
A High-Level Look at the Reranking API
This document explains how the Rerank API endpoint works to perform semantic search by indexing documents based on their relevance to a query.
An Overview of Cohere's Rerank Model
This page describes how Cohere's Rerank models work.
Best Practices for using Rerank
Tips for optimal endpoint performance, including constraints on the number of documents, tokens per document, and tokens per query.
Retrieval Optimization with Rerank
Semantic Search with Embeddings
Examples on how to use the Embed endpoint to perform semantic search (API v1).
Semantic Search with Cohere Embeddings
Introduction to Cohere Embeddings
Text Classification
The document explains how use Cohere's LLM platform to perform text classification tasks.
Text Classification
How to perform text classification using Cohere's classify endpoint.
Introduction to Embeddings at Cohere
Embeddings transform text into numerical data, enabling language-agnostic similarity searches and efficient storage with compression (API v2).
A Guide to Automated Text Classification
The document explains how use Cohere's LLM platform to perform text classification tasks.
Text Classification with Cohere's Classify Endpoint
How to perform text classification using Cohere's classify endpoint.