ChatGPT Pricing 2026: Every Plan, Every Limit Explained

ChatGPT now ships six distinct pricing tiers, and two of those tiers share the "Pro" name at different price points. That naming can mislead buyers. This guide cuts through the confusion, presenting every plan with its exact price, message limits, and the specific user type each tier was designed for.

Whether you're an individual deciding between Free and Plus, a team evaluating Business versus Enterprise, or a developer pricing API calls, every figure here comes from OpenAI's official pricing and announcement pages, verified April 29, 2026.

Explore the broader AI tools hub or start with What Is ChatGPT if you're new to the platform.

ChatGPT pricing spans $0 to $200 per month for consumer and professional subscriptions, plus separate token-based API pricing for developers. There are also two business tiers with per-user pricing.

The Free tier gives access to GPT-5.3 Instant with a hard limit of 10 messages per 5 hours. File uploads are available but restricted. Deep Research is not available on Free. No connectors or integrations are included.

Free is appropriate for occasional casual use and testing the interface before committing to a paid plan.

Go launched on January 15, 2026 as a global budget tier bridging the gap between Free and Plus. It raises the message quota above Free and provides standard chat with basic file handling, but Deep Research remains unavailable.

Go is appropriate for light daily use on a tight budget, and simple Q&A tasks where Deep Research is not a requirement.

What you give up at Go vs Plus: Deep Research access is the decisive gap. If research-quality output matters to your workflow, the $12/month step to Plus changes what the tool can do, not just how fast it does it.

Plus is the primary tier for productive individual use. Its limits as of April 2026:

• GPT-5.4 Thinking: 160 messages per 3 hours
• GPT-5.4 Thinking (weekly): 3,000 messages per week
• Deep Research: 10 sessions per month
• File uploads: 80 files per 3 hours, 20 files per message, 512 MB per file, 25 files per project
• Model access: GPT-5.3 Instant (unlimited), GPT-5.4 Thinking (quota above), GPT-5.4 Pro (limited)

Plus also includes Canvas, Advanced Voice, DALL-E image generation, and standard connectors.

For most knowledge workers, Plus at $20/month covers everyday use. The limit that surfaces most quickly for heavy users is Deep Research: 10 sessions per month runs out fast for researchers or analysts who depend on it daily.

OpenAI launched the Pro $100 tier on April 9, 2026. At $100/month it delivers:

• Deep Research: 500 sessions per month (50x the Plus quota)
• Model quota: 5x the Plus quota across GPT-5.4 Thinking
• Model access: GPT-5.4 Pro at higher quota than Plus
• Priority access to newest models and extended Thinking quotas

Pro $100 is appropriate for heavy professional users who consistently hit Plus message or Deep Research limits. If 10 Deep Research sessions per month leaves you rationing queries, Pro $100 removes that constraint.

Pro $200 is the ceiling of ChatGPT's subscription stack. At $200/month it provides:

• Deep Research: Unlimited sessions
• Message quota: 20x the Plus quota
• Sora video generation: Included
• Context window: GPT-5.4 1M token context window
• Model access: GPT-5.4 Pro at near-unlimited usage, plus operator/computer use priority access
• All features from Pro $100

The 1M token context window and Sora access are the distinguishing capabilities that don't appear at any lower tier. For researchers processing large document collections, or content creators needing both deep analysis and AI video generation in one subscription, Pro $200 is the only consumer tier that covers both.

Pro $200 is appropriate for researchers needing Sora and the 1M context window, and content creators who require both the highest model capability and video generation.

Business is OpenAI's team-level plan. Important constraint: it requires a minimum of 2 users and is not available as a solo subscription.

What Business Includes

• SAML SSO
• SOC 2 Type II certification
• Admin console with usage analytics
• Team workspaces
• Connectors: Google Drive, SharePoint, Slack, GitHub, Notion, Dropbox
• Custom data retention settings
• Model access: GPT-5.3 Instant plus GPT-5.4 Thinking at higher quotas than Plus

Data policy: Customer data is not used for model training. This is the key governance distinction from consumer tiers.

Business is appropriate for teams of 2–149 people who need SSO, data governance controls, and connector sync with existing tooling. If your team is approaching 150 users or requires data residency controls, evaluate Enterprise before committing to Business.

For a comparable per-seat team AI pricing view, see Microsoft Copilot pricing.

Enterprise is available to organizations with 150 or more users. Pricing is negotiated directly with OpenAI's sales team; no published per-seat rate.

What Enterprise Adds Over Business

• 7-region data residency: critical for GDPR and data sovereignty compliance
• Custom fine-tuning
• Dedicated capacity
• Enterprise SLA
• Legal review process
• Enterprise audit logs
• Advanced admin controls
• Custom model deployment
• Model access: GPT-5.4 Pro plus custom model deployment options

The 7-region data residency is the decisive feature for any organization with GDPR obligations or industry data sovereignty rules. That alone separates Enterprise from Business for regulated industries.

Enterprise is appropriate for organizations with 150+ employees that require data residency commitments, custom SLA terms, audit logging for compliance, or fine-tuned model deployments.

The API is a separate cost structure from ChatGPT subscriptions. You pay per million tokens processed, not per message or per month. API access does not include a ChatGPT subscription and vice versa.

Cost Reduction Options

• Cached input: Up to 90% discount on repeated context. The discount depends on cache hit rate; it is not a flat rate.
• Batch API: 50% discount. Requires asynchronous processing with a 24-hour turnaround. Not suitable for real-time applications.

For most production applications, GPT-5.4 Mini at $0.75/$4.50 per million tokens is the default starting point. GPT-5.4 Nano at $0.20/$1.25 fits high-volume low-complexity tasks. GPT-5.4 Pro at $30/$180 is for maximum-capability deployments where cost per call is secondary to output quality.

Developer / API Path

• High-volume, cost-sensitive: GPT-5.4 Mini ($0.75/$4.50/M) or Nano ($0.20/$1.25/M)
• Maximum capability: GPT-5.4 Pro ($30/$180/M)
• Cost reduction: Batch API (50% off, async only) or cached input (up to 90% off repeated context)

For a head-to-head model and pricing comparison, see the ChatGPT vs Claude comparison.