HubSpot Breeze: Features, Pricing, Models & Alternatives (2026)

Last researched: September 26, 2026. Because Breeze pricing is bundled into HubSpot's broader Hub plans and uses a metered credit system, current figures should be reconfirmed at HubSpot's own pricing page (https://www.hubspot.com/pricing) before budgeting. Breeze is the AI product family built by HubSpot, Inc. (NYSE: HUBS) into its Marketing, Sales, Service, Content, and Operations Hubs.

What Is HubSpot Breeze?

Breeze is not a single tool but HubSpot's brand for the AI capabilities layered across its whole customer platform. It's organized into three pillars: Breeze Assistant, a conversational AI panel embedded across HubSpot that can draft content, summarize CRM records, and take actions like updating a deal; Breeze Agents, task-completing AI agents (Content Agent, Prospecting Agent, Customer Agent, Data Agent) that can be extended with a no-code Agent Builder, all managed from a central "Agent Hub"; and Breeze Intelligence, which enriches CRM records with firmographic/contact data, scores leads, and tracks buyer intent signals.

Key Capabilities

• Breeze Assistant sits in a sidebar across HubSpot, understands the page a user is on, and can draft emails, summarize records, or write directly to CRM fields.
• Content Agent generates landing pages, blog drafts, podcasts, and case studies from a prompt and existing brand assets.
• Prospecting Agent (unlocked at the Professional tier and above) researches and sequences outbound prospecting work, metered by credits.
• Customer Agent resolves inbound customer conversations autonomously and, since April 2026, is billed per resolved conversation rather than per attempt.
• Agent Builder is a no-code tool for assembling custom agents from prompts and a business's own CRM data.
• Breeze Intelligence enriches contact/company records, scores leads, and flags buyer-intent signals in near real time.

Main Features

Use Cases

• Drafting marketing and sales content faster from inside the CRM
• Auto-resolving routine customer support conversations
• Enriching and scoring inbound leads without manual research
• Automating early-stage outbound prospecting research
• Building a custom, narrow-purpose agent for a specific internal workflow via Agent Builder

Supported Models / Technologies

HubSpot uses a multi-model strategy rather than a single proprietary model. As of 2026, reporting indicates Breeze Studio agents default to OpenAI's GPT-5 family (including GPT-5.2 and GPT-5.4 Mini/Nano), with Anthropic's Claude models and Google's Gemini 3 Flash / Gemini Flash Image 2.5 used for specific features. HubSpot publishes model-level detail through its AI model cards at trust.hubspot.com/ai; exact model assignment per feature can change and should be checked there.

Platforms

• Web (embedded across all HubSpot Hubs)
• Mobile app (HubSpot mobile app includes Breeze Assistant access)
• Browser extension (HubSpot Sales Chrome extension)
• API (Agent Builder and CRM API access)

Integrations

Native to HubSpot's own CRM, Marketing, Sales, Service, Content, and Commerce Hubs; Slack; Gmail and Outlook; Google Ads and LinkedIn Ads; a large third-party App Marketplace; and an OpenAI ChatGPT connector.

Pricing

Base Hub license cost (Starter/Professional/Enterprise) varies by which Hub(s) and how many seats are purchased; check HubSpot's own pricing page for current Hub license prices, as Breeze credits are additive to that base cost.

Advantages

• Deeply embedded in existing CRM data rather than a bolt-on chatbot
• Outcome-based pricing for Customer Agent means no charge for unresolved attempts
• No-code Agent Builder lets non-developers create narrow custom agents
• Multi-model approach avoids lock-in to a single AI vendor

Limitations

• Credit-metered pricing can make total AI cost harder to predict than a flat fee
• More advanced agents (Prospecting Agent, predictive lead scoring) require Professional or Enterprise Hub tiers
• Some Breeze features moved to outcome-based billing only in April 2026, so older cost estimates online may be outdated
• Full Breeze capability set is tied to being on HubSpot's broader CRM, not usable standalone

Security & Privacy

HubSpot publishes AI model cards and trust documentation at trust.hubspot.com/ai covering which models power which features and their data-handling posture; consult that page and HubSpot's own data-processing agreement for the current customer-data training policy and any enterprise compliance certifications, as specifics can change per feature and were not fully itemized as of this writing.

Best For

Breeze fits organizations already running (or planning to run) their marketing, sales, and support operations inside HubSpot's CRM who want AI drafting, enrichment, and task-agents native to that data, rather than a separate AI tool bolted onto a different CRM.

Alternatives

• Salesforce Agentforce - comparable agent platform, but built around Salesforce's CRM data
• Zoho Zia - AI features bundled into Zoho's CRM suite
• ActiveCampaign AI - AI features inside a marketing-automation-first platform
• Klaviyo AI - AI features focused on e-commerce marketing and messaging
• Adobe Marketo Engage (with Adobe AI features) - enterprise marketing-automation alternative

Frequently Asked Questions

What is HubSpot Breeze?

Breeze is HubSpot's umbrella brand for its AI features, made up of Breeze Assistant (a conversational AI sidebar, formerly called Breeze Copilot), Breeze Agents (pre-built and custom autonomous agents managed through Agent Hub), and Breeze Intelligence (data enrichment, lead scoring, and buyer-intent tracking).

Is Breeze included in HubSpot's price, or is it extra?

Agent Hub and core Breeze features are included with HubSpot's Starter, Professional, and Enterprise Hub plans, but usage-metered features — like Breeze Intelligence enrichment or the Customer Agent — draw down a pool of included credits and can incur additional pay-as-you-go or credit-pack costs once that allowance is used.

What AI models power HubSpot Breeze?

HubSpot uses a multi-model approach rather than one proprietary model: reporting as of 2026 points to OpenAI's GPT-5 family (GPT-5, GPT-5.2, GPT-5.4 Mini/Nano) as the default for Breeze Studio agents, alongside Anthropic's Claude models and Google's Gemini 3 Flash for certain features; HubSpot publishes model details via its AI model cards at trust.hubspot.com/ai.

How does Breeze Customer Agent pricing work?

Since April 14, 2026, HubSpot moved Breeze Customer Agent and Breeze Prospecting Agent to outcome-based pricing, meaning customers are charged in credits only when the agent completes its task — for example, 50 credits (about $0.50) per resolved conversation for the Customer Agent — rather than being charged regardless of outcome.

Final Summary

Breeze's value is inseparable from HubSpot's underlying CRM data, and its credit-metered, outcome-based pricing model (updated as recently as April 2026) means the real monthly cost depends heavily on usage volume, not just the Hub tier purchased. Teams should model expected conversation and enrichment volume against the credit-cost table above rather than budgeting a flat "AI" line item.