North

What it does

North is Cohere's enterprise AI platform for agents, search and workflow automation over company knowledge. It works from documents and data in connected systems, and produces grounded answers, summaries, tables, charts, decks and small apps inside a chat-based workspace.

Teams use it to build agents that draw on shared libraries, reusable skills and persistent memory across sessions. They can also describe an automation in natural language and monitor it with built-in analytics. Pre-built data connectors, flexible APIs and built-in connectors link North to existing tools, data and monitoring systems.

North is aimed at enterprise teams in legal, sales, finance and operations, especially where security and data control matter. Guardrails govern how users interact with AI, and an admin panel manages users and agents. It can run in your own VPC, on-premises, or through Cohere's Model Vault inference platform. Access starts with an instant demo or a conversation with sales. Pricing is custom, so request a quote and confirm connectors and deployment scope with Cohere.

What it can do

• Agents grounded in an organization's knowledge sources
• Natural-language automations with monitoring analytics
• Reusable skills that agents can call
• Shared libraries of knowledge and assets
• Generation of documents, dashboards, decks and apps from chat
• Persistent memory across sessions
• Guardrails governing user interaction with AI
• Admin controls for users and agents

Works with

• Pre-built data connectors
• Flexible APIs and built-in connectors to existing tools, data and monitoring systems

Example workflows

Contract review assistance for legal teams

Input: contract documents and legal reference material in connected repositories. Steps: an agent searches the knowledge base, a skill applies review logic, and guardrails govern user access. Output: grounded summaries and redline suggestions for legal review.

Recurring reporting automation for finance or operations

Input: data sources connected to North and a natural-language description of the report. Steps: build an automation from the description, schedule it, and monitor it with analytics. Output: generated tables, charts and a summary document shared through a library.

Before you choose

• North pricing is custom enterprise pricing; request a quote from Cohere sales.
• Model Vault is priced per instance by model and performance tier, separately from North itself, so confirm which your deployment requires.
• Private deployment in a VPC or on-premises depends on your infrastructure and should be scoped with Cohere sales.

Listed under

• Operations
• Workflow automation
• Agents
• AI agents
• Business teams
• Enterprise
• Enterprise AI
• Enterprise search

Get started

The page offers an instant demo and a sales contact route; sign-in is through the Cohere dashboard.

1. Sign in to the Cohere dashboard or request a demo through the North demo page.
2. Connect your organization's data sources using the pre-built connectors or APIs.
3. Configure admin settings, guardrails and shared libraries for users and agents.

North requires an enterprise quote from Cohere. API usage and Model Vault rates are separate products; confirm deployment, support and usage terms for your North contract.

North doesn’t publish prices. You get a quote by contacting sales. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

• North 2 | Enterprise AI Without Compromises | Cohere ↗cohere.com
• Pricing | Secure and Scalable Enterprise AI | Cohere ↗cohere.com
• Enterprise Open Source AI Models | Cohere ↗cohere.com

Company website

cohere.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.