How Much Does an AI Chatbot Cost in 2026? A Pricing Guide

TL;DR

• A working AI chatbot in 2026 costs anywhere from $0 to several hundred dollars per month, depending on the pricing model.
• Flat-rate plans (like Chatonbo) charge a fixed monthly fee regardless of usage: Free at $0, Pro at $19/mo, Business at $59/mo. No per-conversation or per-resolution fees.
• Usage-based plans (like Intercom Fin or Zendesk AI) charge per resolution or per seat - often around $0.99 per resolved conversation plus a per-agent seat fee. Costs scale with volume and can be unpredictable.
• What drives cost: conversation volume, pricing model, number of human seats, and whether "agentic" actions (order tracking, recommendations) cost extra.
• Rule of thumb: if you handle predictable or high volume, flat pricing is cheaper. If you handle a low, spiky volume, per-resolution can work - but model it first.

How much does an AI chatbot cost?

An AI chatbot costs between $0 and $59 per month on flat-rate plans, or roughly $0.99 per resolved conversation plus a seat fee on usage-based plans. The exact number depends almost entirely on which of two pricing models a vendor uses - and how much traffic your site gets.

That's the whole answer in one line. The rest of this guide explains the two models, what actually drives your bill, and how to estimate your own cost before you commit to anything.

The two AI chatbot pricing models

Nearly every AI chatbot on the market in 2026 charges one of two ways:

1. Flat monthly pricing

You pay a fixed fee per month and get a generous conversation allowance. Whether you handle 500 chats or 5,000, the price doesn't move. This is predictable, easy to budget, and friendly to growth - busy months don't trigger surprise invoices.

[Chatonbo](/pricing) uses flat pricing:

• Free - $0/mo: 100 conversations/month, 1 bot, 3 knowledge sources
• Pro - $19/mo: 5,000 conversations/month
• Business - $59/mo: 10,000 conversations/month
• Enterprise - custom

There are no per-conversation or per-resolution fees. Agentic features - order tracking, product recommendations, discount validation, and one-click checkout links - are included with no extra per-use charge.

2. Usage-based pricing (per resolution or per seat)

You pay for outcomes or for people. Two common variants:

• Per resolution: you're billed each time the bot "resolves" a conversation. Pricing is listed at around $0.99 per resolution on some platforms.
• Per seat: you pay a monthly fee for every human agent who has access, often stacked on top of resolution fees.

Usage-based pricing can look cheap at low volume. The risk is that costs scale directly with your traffic - and a viral week, a product launch, or a holiday rush can multiply your bill.

AI chatbot pricing models compared

The table below compares the common pricing models in 2026. Competitor figures are illustrative, based on publicly listed pricing, and subject to change - always confirm current rates with each vendor.

*Pricing for tools other than Chatonbo is illustrative and listed publicly at the time of writing; verify before purchasing.*

What actually drives your AI chatbot cost

Four variables decide your real bill:

1. Conversation volume. The single biggest factor. On flat plans it determines which tier you need. On usage plans it directly multiplies your cost.
2. Pricing model. Flat vs per-resolution can produce wildly different bills for the *same* traffic. We'll model this below.
3. Number of human seats. Per-seat platforms get expensive as your support team grows. Flat AI-first tools don't charge per agent.
4. "Agentic" actions. Some platforms charge extra when the bot *does* something - looks up an order, applies a discount, recommends a product. On Chatonbo these [agentic commerce](/glossary/agentic-commerce) actions are included at no per-use charge.

A hidden fifth factor: the underlying AI model. Chatonbo runs on GPT-5.5, Claude, or Gemini (selectable per bot), supports 50+ languages, and works on Shopify, WooCommerce, Wix, WordPress, or any website - all within the same flat plan. You don't pay more to switch models or add a language.

How to estimate your own AI chatbot cost

You can ballpark your cost in three steps:

Step 1 - Estimate your monthly conversations

Check your analytics. A useful starting point: 1–3% of monthly website visitors will start a chat. So 10,000 visitors ≈ 100–300 conversations. If you already run live chat or a help desk, use last month's actual ticket count.

Step 2 - Run both models

Take your estimated volume and price it both ways. For example, 2,000 conversations/month:

• Flat (Chatonbo Pro): $19/mo, full stop - you're well under the 5,000 cap.
• Per resolution (illustrative, ~$0.99 each): if even half resolve via the bot, that's ~1,000 × $0.99 = ~$990/mo, before any seat fees.

The gap is the point. At meaningful volume, flat pricing is usually dramatically cheaper. At very low volume - say 30 conversations a month - per-resolution might edge out a paid plan, though a free flat tier often wins outright.

Step 3 - Add seats and extras

If a platform charges per seat, multiply by your number of agents. If it charges for agentic actions, estimate how often the bot looks up orders or applies discounts, and add that in. Flat, AI-first plans skip both of these line items.

AI chatbot cost FAQ

Is there a free AI chatbot?

Yes. Chatonbo's Free plan is $0/mo and includes 100 conversations per month, 1 bot, and 3 knowledge sources - enough to launch and test on a real website.

Why is per-resolution pricing risky?

Because your cost scales with traffic. A successful campaign or seasonal spike increases conversations - and your bill grows in lockstep, sometimes unpredictably.

What does "resolution" mean in chatbot pricing?

A resolution is a conversation the AI is deemed to have handled or closed without a human. Definitions vary by vendor, which makes usage-based bills harder to forecast.

Do agentic features cost extra?

On some platforms, yes. On Chatonbo, agentic actions like order tracking, product recommendations, discount validation, and checkout links are included in the flat plan with no per-use charge. See the [AI shopping assistant](/solutions/ai-shopping-assistant) for what that looks like in practice.

Flat vs usage-based - which is cheaper?

For predictable or growing volume, flat pricing almost always wins. For very low, spiky volume, usage-based *can* be cheaper - but model both before deciding. For a side-by-side on one popular usage-based tool, see [Chatonbo vs Intercom](/compare/intercom).

The bottom line

The real cost of an AI chatbot in 2026 isn't a single number - it's a function of your volume and your vendor's pricing model. Flat pricing trades a small fixed fee for total predictability. Usage-based pricing trades a low entry point for bills that move with your traffic.

If you want a bot you can budget for - one fixed price, agentic features included, no per-conversation surprises - [start free with Chatonbo](/platform/register) and upgrade only when your volume actually calls for it.