Build customer-facing agents and a help center, test them in Studio, and route conversations to a team when needed. New-workspace pricing is based on conversations, with separate AI spend and automatic paid top-ups.
Botpress positions its current platform around customer support: AI agents, a public Help Center, the Studio builder and Desk for human handoff. A company can begin with a narrow support topic, connect approved knowledge, test the agent's answers and publish a webchat experience. The product is more than a scripted FAQ widget because a conversation can involve knowledge retrieval, an AI model and a human support desk. This increases the need for good source content and clear escalation rules.
Use a small pilot before placing the widget on a live checkout or account page. Import a set of current, owned support articles; build answers for five common questions and three deliberately ambiguous ones. Check whether the agent cites the correct policy version, refuses to invent a refund promise, and transfers cases requiring account access to a person. Then test a conversation that spans a month boundary, because the current pricing FAQ says it counts in both months. Monitor unknown answers, containment and handoff quality, not only the number of chats.
Botpress changed its pricing for workspaces created after May 14, 2026; legacy workspaces may remain on older plans. The current page lists Free with 100 conversations, Plus at $150 per month billed annually with 250 monthly conversations, and Team at $750 per month billed annually with 1,500. A conversation is defined by at least two end-user messages during a billing month; both bot and human handling count. Paid tiers include some AI spend, but the FAQ describes automatic purchases of conversation packs and AI credit grants near quota. Budget both meters and understand the overage behavior before production traffic.
Product FAQ agent
Order-support triage
Internal support desk
Official product image. Click to inspect the details.
Define flows, knowledge and replies, then use the emulator to test ambiguous inputs before deployment.
Give customers approved self-service articles and make sure agent answers follow the same current policy.
Route complex or account-specific cases to a person with enough context to continue efficiently.
Use supported webchat and messaging channels where the plan allows, while checking channel-specific setup and consent.
Track conversations, AI spend, seats, routing and role-based access as the support operation grows.
Choose one support topic
Build and test in Studio
Pilot with human coverage
Budget and review monthly
Select a high-volume, low-risk question group and collect current approved answers plus escalation rules.
Create the agent and Help Center, then test normal, ambiguous and out-of-scope conversations.
Publish to a limited audience, inspect transcripts and handoffs, and correct misleading answers quickly.
Track conversation count and AI spend separately, including automatic top-ups and cross-month conversations.
Pricing checked on October 7, 2026. See plan details below.
The current Free tier lists 100 conversations and three seats/AI agents, with no top-ups or overages. Plus lists $150 per month when billed annually and 250 conversations per month; Team lists $750 per month billed annually and 1,500 conversations. The May 2026 announcement says this pricing applies to workspaces created after that date, while legacy workspaces may have different terms. Check the account creation date and billing toggle before comparing plans.
Conversation counts are separate from AI model spend. The pricing FAQ says paid tiers automatically add 100-conversation packs near quota and purchase $10 AI spend grants at a threshold; the page lists $25 AI usage included in Plus and $150 in Team. Packs and grants have different expiry rules. Because a cross-month or human-assisted exchange can also count, a production budget should be based on measured support traffic and real model use, not only resolved tickets.
The pricing FAQ counts an exchange with at least two end-user messages during a billing month.
Yes. The current definition counts AI-only and human-assisted conversations.
It is a separate AI-spend meter; paid tiers include some allowance and may auto-purchase grants.
The current FAQ says paid-plan auto-recharge cannot be turned off; review that behavior before launch.
The announcement says new workspaces receive the new pricing; existing legacy workspaces remain on old plans for now.
Test correct policy answers, unknown questions, escalation, a cross-month thread and the usage counters.
Explore platforms, inputs and outputs, licensing, and access requirements.
Botpress pricing defines a billable conversation by at least two end-user messages in a billing month, even when a human handles it; a cross-month thread can count in both months. Its Free plan has a hard cap. Paid plans may auto-add capacity, so usage monitoring and an operational owner are part of the setup.
Support conversations can include personal data, orders and account problems. Restrict knowledge sources to approved content, avoid exposing account details in public Help Center pages, and define handoff and retention policy before launch. Test an agent response against real policy and require human handling for sensitive decisions.
Current Botpress product platform ↗
Current plan and conversation billing FAQ ↗
Loading experiences…
Saved tools are private. Approved comments are public; edits return to moderation. Comment counts include approved comments and replies. Each person has one rating. New ratings are automatically approved; withdrawn, unapproved or invalid ratings do not count. Share real experiences and avoid spam, private information or personal attacks.
For moderation appeals or data requests: support@findgoodai.com
Character.AI is a conversational entertainment platform built around fictional personas. Its appeal is sustained dialogue, character discovery and collaborative storytelling rather than authoritative answers to factual questions.
Bring files, knowledge and AI tools together for research, drafting and recurring work. TextCortex provides a free entry and paid model usage, with custom agents and connected knowledge features.
Design, test and deploy an agent across web chat or voice, grounding answers in an approved knowledge base and routing exceptions to people. The current public pricing page offers a free trial and usage-based or quoted plans, without a stable numeric self-serve table.
Bring a real task and see how it fits the way you work.
User experiences