Cohere
Explore the Cohere platform and its large language models accessible via API, focusing on how to integrate generative and representation models into Go applications. Understand model types like Command and Embed, and learn to use Cohere's API endpoints for tasks such as classification, summarization, and chat. This lesson guides you through setting up the Cohere API and building a classification example with Go code and the Go client.
Introduction to Cohere
Cohere platform serves large language models (LLM) as a fully-managed service. Since these LLMs are accessible via an API, users do not need to handle the complexities of deploying, operating, and scaling LLM infrastructure.
The Cohere platform provides two types of models:
Generative models: They generate text in response to a text input (or prompt). This is useful for question answering, copywriting, text summarization, and similar use cases.
Representative models: They take text input and generate an embedding, which is a numerical value that captures its contextual meaning. This is useful for semantic search, text classification, topic modeling, and similar use cases.
Command — Cohere generative model
The Command model can take a user's instruction and generate text following the instruction. Since it has conversational capabilities, it is also suitable for chat applications. There are two categories of the Command model:
Command: It enables use cases such as copy generation, chat, text summarization, etc. It should be used for ...