GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • docs
  • DASHBOARD
  • PLAYGROUND
  • COMMUNITY
  • LOG IN
  • Guides and concepts
  • API Reference
  • Release Notes
  • LLMU
  • Cookbooks
  • An Overview of The Cohere PlatformNext
  • Installation
  • Creating a client
  • Playground

This view

  • Original page
  • New address

GeezerVison is showing docs.cohere.com — Cohere Documentation

Skip to navigation
On this page
  • Cohere products
  • Cohere Platform
  • Model Vault
  • North
  • Which should I use?
This page is for version v2 API (default). For other versions, use one of these documentation indexes: - v2 API (default): https://docs.cohere.com/v2/llms.txt - v1 API: https://docs.cohere.com/v1/llms.txt

Cohere builds world-class AI models:

  • Command for chat, RAG, tool use, and agents.
  • North Mini Code for agentic coding and software engineering tasks.
  • Embed for semantic search, classification, and clustering.
  • Rerank for improving search relevance.

You can put these models to work through any of Cohere’s products. This page helps you find the right one for what you’re building.

Cohere products

Every product is powered by Cohere’s models but meets a different need: from calling the model APIs directly, to running them in a dedicated, isolated deployment, to adopting a ready-made AI assistant that does the work for you.

Cohere Platform

Best for getting started fast. The Cohere Platform lets you call our models directly over the API on Cohere’s cloud, with no infrastructure to manage. Grab an API key and start building in minutes.

  • New here? Read the platform overview.
  • Install an SDK (Python, TypeScript, Java, or Go) and create a client.
  • Explore the models and the API reference.

Model Vault

Best for security, isolation, and control. Model Vault deploys Cohere’s models and leading open models, such as DeepSeek, Gemma, GPT-OSS, and Qwen, on dedicated infrastructure in the United States, the European Union, or Canada, so your workloads run in isolation rather than on shared infrastructure. You call it over the same API, without operating the serving stack yourself. For regulated or highly sensitive workloads, Model Vault Encrypted adds confidential computing, so you can run fully encrypted AI inference inside hardware-secured enclaves and cryptographically verify that protection at any time through remote attestation.

Explore Model Vault →

North

Best for a ready-to-use AI solution. North is Cohere’s agentic AI platform: a complete application that puts the models to work for your team out of the box. Chat with a helpful AI assistant, ground answers in your connected documents and data, create and refine documents, and build customizable agents and workflow automations, with secure integrations to systems like Google Drive, SharePoint, Slack, Salesforce, and MCP servers. North can be deployed and run in your own environment (on-premises or air-gapped), with strong security, data, and compliance controls.

Read the North documentation →

Which should I use?

If you want to…Use
Get started fast with direct API access on shared infrastructureCohere Platform
Deploy Cohere’s models or leading open models securely in a dedicated, isolated environmentModel Vault
Run fully encrypted, verifiably private inference for your most sensitive or regulated workloadsModel Vault Encrypted
Adopt a ready-made agentic AI assistant that works over your internal dataNorth
You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.