Back
Cohere
Language Models (GPT) - The leading provider of conversational AI and language processing

Heading 1

Heading 2

Heading 3

Heading 4

Heading 5
Heading 6

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.

Block quote

Ordered list

  1. Item 1
  2. Item 2
  3. Item 3

Unordered list

  • Item A
  • Item B
  • Item C

Text link

Bold text

Emphasis

Superscript

Subscript

<ul><li><strong>Pricing Model:</strong> Usage-based</li><li><strong>Packaging Model:</strong> Good / Better / Best</li><li><strong>Credit Model:</strong> Not used</li></ul>
December 17, 2025
Last update:
<h3>Product Overview</h3><p>Cohere is an enterprise AI platform specializing in large language models and AI infrastructure for businesses. Founded in 2019 by Aidan Gomez (co-author of the transformer architecture paper), the company focuses on secure, enterprise-grade AI deployment with features like air-gapped installations and private cloud options. Cohere&#039;s product suite includes generative AI models (Command series), embedding models (Embed), and reranking services (Rerank), all designed for business applications requiring high security and compliance standards.<br /> <br /> Cohere&#039;s transparent, usage-based pricing model has been instrumental in its rapid market traction, enabling the company to reach $100M in annualized revenue as of May 2025. By offering pure consumption-based billing without upfront commitments, Cohere lowered adoption barriers for enterprises experimenting with AI while capturing expansion revenue as production workloads scaled. This monetization approach, combined with $1.6B in total funding at a $7B valuation, has positioned Cohere as a leader in enterprise AI. The company differentiates itself through enterprise-focused features including on-premises deployment options, multilingual capabilities across 23 languages, and industry-leading context windows up to 256K tokens.</p>
<h3>Pricing Snapshot</h3><div class="tableResponsive"><table cellpadding="6" cellspacing="0"><tr><th>Model</th><th>Input Price</th><th>Output Price</th><th>Context Window</th><th>Status</th></tr><tr><td>Command R7B</td><td>$0.0375/1M tokens</td><td>$0.15/1M tokens</td><td>128K tokens</td><td>Production</td></tr><tr><td>Command R</td><td>$0.15/1M tokens</td><td>$0.60/1M tokens</td><td>128K tokens</td><td>Production</td></tr><tr><td>Aya Expanse</td><td>$0.50/1M tokens</td><td>$1.50/1M tokens</td><td>128K tokens</td><td>Production</td></tr><tr><td>Command R+</td><td>$2.50/1M tokens</td><td>$10.00/1M tokens</td><td>128K tokens</td><td>Production</td></tr><tr><td>Command A</td><td>$2.50/1M tokens</td><td>$10.00/1M tokens</td><td>256K tokens</td><td>Production</td></tr><tr><td>Embed 4</td><td>$0.12/1M tokens</td><td>-</td><td>128K tokens</td><td>Production</td></tr><tr><td>Rerank 3.5</td><td>$2.00/1K searches</td><td>-</td><td>100 docs/request</td><td>Production</td></tr></table></div>
<h3>Key Features & Capabilities</h3><p>Cohere offers a complete enterprise AI stack including generative models (Command family), embeddings (Embed), and relevance ranking (Rerank), with flexible deployment options spanning API access, multi-cloud integrations, and private cloud installations.</p><ul><li>Command Model Family: Range from Command R7B for cost-optimized operations to Command A for premium enterprise applications with 256K context windows</li><li>Embed &amp; Rerank Models: Enhanced accuracy for search, classification, clustering, RAG implementations, and relevance ordering</li><li>Flexible Deployment: Direct API access, multi-cloud integrations (Oracle, AWS, Google Cloud, Azure), and private cloud/on-premises options</li><li>Enterprise-Grade Controls: Data privacy controls, context windows up to 256K tokens, production rate limits of 60-120 requests/minute</li></ul>
<h3>Pricing Model Analysis</h3><p>Cohere operates on a transparent pay-as-you-go usage-based model with à la carte services that customers can select:</p><div class="tableResponsive"><table cellpadding="6" cellspacing="0"><tr><th>Metric Type</th><th>What Measured</th><th>Why It Matters</th></tr><tr><td>Value Metric</td><td>Tokens processed (input/output split)</td><td>Aligns cost with computational resources used</td></tr><tr><td>Usage Metric</td><td>API calls, context window utilization</td><td>Drives billing based on actual consumption</td></tr><tr><td>Billable Metric</td><td>Monthly usage OR $250 threshold</td><td>Enables flexible billing cycles for different usage patterns</td></tr></table></div>
<h3>Pricing Evolution Timeline</h3><div class="tableResponsive"><table cellpadding="6" cellspacing="0"><tr><th>Date</th><th>Milestone</th><th>Source</th></tr><tr><td>October 18, 2022</td><td>Free Developer Tier Launch with pay-as-you-go production access</td><td><a href='https://cohere.com/blog/free-developer-tier-announcement' target='_blank'>Cohere Blog <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 16 16" fill="none"><path d="M14 6.5C14 6.63261 13.9473 6.75979 13.8536 6.85355C13.7598 6.94732 13.6326 7 13.5 7C13.3674 7 13.2402 6.94732 13.1464 6.85355C13.0527 6.75979 13 6.63261 13 6.5V3.7075L8.85437 7.85375C8.76055 7.94757 8.63331 8.00028 8.50062 8.00028C8.36794 8.00028 8.2407 7.94757 8.14688 7.85375C8.05305 7.75993 8.00035 7.63268 8.00035 7.5C8.00035 7.36732 8.05305 7.24007 8.14688 7.14625L12.2925 3H9.5C9.36739 3 9.24021 2.94732 9.14645 2.85355C9.05268 2.75979 9 2.63261 9 2.5C9 2.36739 9.05268 2.24021 9.14645 2.14645C9.24021 2.05268 9.36739 2 9.5 2H13.5C13.6326 2 13.7598 2.05268 13.8536 2.14645C13.9473 2.24021 14 2.36739 14 2.5V6.5ZM11.5 8C11.3674 8 11.2402 8.05268 11.1464 8.14645C11.0527 8.24021 11 8.36739 11 8.5V13H3V5H7.5C7.63261 5 7.75979 4.94732 7.85355 4.85355C7.94732 4.75979 8 4.63261 8 4.5C8 4.36739 7.94732 4.24021 7.85355 4.14645C7.75979 4.05268 7.63261 4 7.5 4H3C2.73478 4 2.48043 4.10536 2.29289 4.29289C2.10536 4.48043 2 4.73478 2 5V13C2 13.2652 2.10536 13.5196 2.29289 13.7071C2.48043 13.8946 2.73478 14 3 14H11C11.2652 14 11.5196 13.8946 11.7071 13.7071C11.8946 13.5196 12 13.2652 12 13V8.5C12 8.36739 11.9473 8.24021 11.8536 8.14645C11.7598 8.05268 11.6326 8 11.5 8Z" fill="#95988B"/></svg></a></td></tr><tr><td>April 4, 2024</td><td>Command R+ Premium Model Launch at $3.00/$15.00 per 1M tokens</td><td><a href='https://cohere.com/blog/command-r-plus-microsoft-azure' target='_blank'>Cohere Blog <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 16 16" fill="none"> <path d="M14 6.5C14 6.63261 13.9473 6.75979 13.8536 6.85355C13.7598 6.94732 13.6326 7 13.5 7C13.3674 7 13.2402 6.94732 13.1464 6.85355C13.0527 6.75979 13 6.63261 13 6.5V3.7075L8.85437 7.85375C8.76055 7.94757 8.63331 8.00028 8.50062 8.00028C8.36794 8.00028 8.2407 7.94757 8.14688 7.85375C8.05305 7.75993 8.00035 7.63268 8.00035 7.5C8.00035 7.36732 8.05305 7.24007 8.14688 7.14625L12.2925 3H9.5C9.36739 3 9.24021 2.94732 9.14645 2.85355C9.05268 2.75979 9 2.63261 9 2.5C9 2.36739 9.05268 2.24021 9.14645 2.14645C9.24021 2.05268 9.36739 2 9.5 2H13.5C13.6326 2 13.7598 2.05268 13.8536 2.14645C13.9473 2.24021 14 2.36739 14 2.5V6.5ZM11.5 8C11.3674 8 11.2402 8.05268 11.1464 8.14645C11.0527 8.24021 11 8.36739 11 8.5V13H3V5H7.5C7.63261 5 7.75979 4.94732 7.85355 4.85355C7.94732 4.75979 8 4.63261 8 4.5C8 4.36739 7.94732 4.24021 7.85355 4.14645C7.75979 4.05268 7.63261 4 7.5 4H3C2.73478 4 2.48043 4.10536 2.29289 4.29289C2.10536 4.48043 2 4.73478 2 5V13C2 13.2652 2.10536 13.5196 2.29289 13.7071C2.48043 13.8946 2.73478 14 3 14H11C11.2652 14 11.5196 13.8946 11.7071 13.7071C11.8946 13.5196 12 13.2652 12 13V8.5C12 8.36739 11.9473 8.24021 11.8536 8.14645C11.7598 8.05268 11.6326 8 11.5 8Z" fill="#95988B"/></svg></a></td></tr><tr><td>June 10, 2024</td><td>Billing threshold system implemented ($150 warning, $250 auto-charge)</td><td><a href='https://docs.cohere.com/v2/changelog/release-notes-for-june-10th-2024#changes-to-billing' target='_blank'>Cohere Release Notes <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 16 16" fill="none"><path d="M14 6.5C14 6.63261 13.9473 6.75979 13.8536 6.85355C13.7598 6.94732 13.6326 7 13.5 7C13.3674 7 13.2402 6.94732 13.1464 6.85355C13.0527 6.75979 13 6.63261 13 6.5V3.7075L8.85437 7.85375C8.76055 7.94757 8.63331 8.00028 8.50062 8.00028C8.36794 8.00028 8.2407 7.94757 8.14688 7.85375C8.05305 7.75993 8.00035 7.63268 8.00035 7.5C8.00035 7.36732 8.05305 7.24007 8.14688 7.14625L12.2925 3H9.5C9.36739 3 9.24021 2.94732 9.14645 2.85355C9.05268 2.75979 9 2.63261 9 2.5C9 2.36739 9.05268 2.24021 9.14645 2.14645C9.24021 2.05268 9.36739 2 9.5 2H13.5C13.6326 2 13.7598 2.05268 13.8536 2.14645C13.9473 2.24021 14 2.36739 14 2.5V6.5ZM11.5 8C11.3674 8 11.2402 8.05268 11.1464 8.14645C11.0527 8.24021 11 8.36739 11 8.5V13H3V5H7.5C7.63261 5 7.75979 4.94732 7.85355 4.85355C7.94732 4.75979 8 4.63261 8 4.5C8 4.36739 7.94732 4.24021 7.85355 4.14645C7.75979 4.05268 7.63261 4 7.5 4H3C2.73478 4 2.48043 4.10536 2.29289 4.29289C2.10536 4.48043 2 4.73478 2 5V13C2 13.2652 2.10536 13.5196 2.29289 13.7071C2.48043 13.8946 2.73478 14 3 14H11C11.2652 14 11.5196 13.8946 11.7071 13.7071C11.8946 13.5196 12 13.2652 12 13V8.5C12 8.36739 11.9473 8.24021 11.8536 8.14645C11.7598 8.05268 11.6326 8 11.5 8Z" fill="#95988B"/></svg></a></td></tr><tr><td>August 30, 2024</td><td>Command R+ price reduction to $2.50/$10.00 per 1M tokens (33% output price cut)</td><td><a href='https://docs.cohere.com/changelog/command-gets-refreshed' target='_blank'>Cohere Changelog <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 16 16" fill="none"><path d="M14 6.5C14 6.63261 13.9473 6.75979 13.8536 6.85355C13.7598 6.94732 13.6326 7 13.5 7C13.3674 7 13.2402 6.94732 13.1464 6.85355C13.0527 6.75979 13 6.63261 13 6.5V3.7075L8.85437 7.85375C8.76055 7.94757 8.63331 8.00028 8.50062 8.00028C8.36794 8.00028 8.2407 7.94757 8.14688 7.85375C8.05305 7.75993 8.00035 7.63268 8.00035 7.5C8.00035 7.36732 8.05305 7.24007 8.14688 7.14625L12.2925 3H9.5C9.36739 3 9.24021 2.94732 9.14645 2.85355C9.05268 2.75979 9 2.63261 9 2.5C9 2.36739 9.05268 2.24021 9.14645 2.14645C9.24021 2.05268 9.36739 2 9.5 2H13.5C13.6326 2 13.7598 2.05268 13.8536 2.14645C13.9473 2.24021 14 2.36739 14 2.5V6.5ZM11.5 8C11.3674 8 11.2402 8.05268 11.1464 8.14645C11.0527 8.24021 11 8.36739 11 8.5V13H3V5H7.5C7.63261 5 7.75979 4.94732 7.85355 4.85355C7.94732 4.75979 8 4.63261 8 4.5C8 4.36739 7.94732 4.24021 7.85355 4.14645C7.75979 4.05268 7.63261 4 7.5 4H3C2.73478 4 2.48043 4.10536 2.29289 4.29289C2.10536 4.48043 2 4.73478 2 5V13C2 13.2652 2.10536 13.5196 2.29289 13.7071C2.48043 13.8946 2.73478 14 3 14H11C11.2652 14 11.5196 13.8946 11.7071 13.7071C11.8946 13.5196 12 13.2652 12 13V8.5C12 8.36739 11.9473 8.24021 11.8536 8.14645C11.7598 8.05268 11.6326 8 11.5 8Z" fill="#95988B"/></svg></a></td></tr><tr><td>August 6, 2025</td><td>North Enterprise Platform GA with custom pricing model</td><td><a href='https://cohere.com/blog/north-ga' target='_blank'>Cohere Blog <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 16 16" fill="none"><path d="M14 6.5C14 6.63261 13.9473 6.75979 13.8536 6.85355C13.7598 6.94732 13.6326 7 13.5 7C13.3674 7 13.2402 6.94732 13.1464 6.85355C13.0527 6.75979 13 6.63261 13 6.5V3.7075L8.85437 7.85375C8.76055 7.94757 8.63331 8.00028 8.50062 8.00028C8.36794 8.00028 8.2407 7.94757 8.14688 7.85375C8.05305 7.75993 8.00035 7.63268 8.00035 7.5C8.00035 7.36732 8.05305 7.24007 8.14688 7.14625L12.2925 3H9.5C9.36739 3 9.24021 2.94732 9.14645 2.85355C9.05268 2.75979 9 2.63261 9 2.5C9 2.36739 9.05268 2.24021 9.14645 2.14645C9.24021 2.05268 9.36739 2 9.5 2H13.5C13.6326 2 13.7598 2.05268 13.8536 2.14645C13.9473 2.24021 14 2.36739 14 2.5V6.5ZM11.5 8C11.3674 8 11.2402 8.05268 11.1464 8.14645C11.0527 8.24021 11 8.36739 11 8.5V13H3V5H7.5C7.63261 5 7.75979 4.94732 7.85355 4.85355C7.94732 4.75979 8 4.63261 8 4.5C8 4.36739 7.94732 4.24021 7.85355 4.14645C7.75979 4.05268 7.63261 4 7.5 4H3C2.73478 4 2.48043 4.10536 2.29289 4.29289C2.10536 4.48043 2 4.73478 2 5V13C2 13.2652 2.10536 13.5196 2.29289 13.7071C2.48043 13.8946 2.73478 14 3 14H11C11.2652 14 11.5196 13.8946 11.7071 13.7071C11.8946 13.5196 12 13.2652 12 13V8.5C12 8.36739 11.9473 8.24021 11.8536 8.14645C11.7598 8.05268 11.6326 8 11.5 8Z" fill="#95988B"/></svg></a></td></tr></table></div>
<h3>Customer Sentiment Highlights</h3><ul><li>“We are blown away by Cohere Embed 4&#039;s ability to accurately surface relevant products to search queries... Being able to represent our products in a unified embedding makes our search faster and our internal tooling more efficient.” — <b>Enterprise Customer, Product Hunt</b></li><li>“We used Cohere as it was a cost-effective document indexing solution that provided efficient embedding generation for knowledge base features.” — <b>Enterprise Customer, Product Hunt</b></li></ul>
Metronome’s Take
<p>Cohere implements pure usage-based pricing with token-level metering across their model portfolio, creating transparent alignment between computational resources consumed and customer costs. This infrastructure-style pricing serves production AI workloads where businesses need precise cost attribution and can absorb variable billing patterns.</p>
<p><strong>Recommendation:</strong> Organizations operating production AI systems with sophisticated cost management benefit most from this type of pricing model. Companies requiring fixed monthly budgets or serving price-sensitive end users might look to utilize usage caps or prepaid credit options.</p>
<h4>Key Insights</h4><ul><li><strong>Tiered Model Economics:</strong> A 4x output-to-input token multiplier ($0.15 vs $0.0375 on Command R7B) reflects generation complexity while providing clear cost signals <p><strong>Benefit:</strong> Customers optimizing prompts can dramatically reduce bills by minimizing output verbosity, creating natural efficiency incentives.</p></li><li><strong>Enterprise Deployment Flexibility:</strong> Private deployment options (VPC, on-premises) complement API pricing. <p><strong>Benefit:</strong> This consumption model scales from prototyping through production across deployment architectures by addressing data residency requirements without forcing customers into rigid annual commits</p></li></ul>

The Pricing
Experimentation
Playbook

Find your ideal pricing model

Answer 8 quick questions to discover which best fits how your customers get value from your product.

Find your model