Chatbase
No-code platform for building and deploying AI support agents trained on your own business data
AI agent and chatbot platform with a visual studio, a TypeScript ADK, and LLM usage bundled into per-conversation pricing
Botpress is a cloud platform for building AI agents and support chatbots, now centred on an AI-native helpdesk. Pricing is per conversation, not per seat: the free tier covers 100 conversations with 3 seats, Plus is $150/mo (250 conversations), and Team is $750/mo (1,500 conversations, unlimited seats), all billed annually. LLM inference is bundled — there is no separate token bill. The open-source v12 self-hosted edition is sunset; cloud only.
Botpress is a cloud platform for building AI agents that handle customer conversations. It started as an open-source chatbot framework and has spent roughly a decade as developer infrastructure, but the current product is an enterprise support platform: Botpress Desk is an AI-native helpdesk, Botpress Studio is a visual agent builder, and the Agent Development Kit (ADK) is a TypeScript CLI framework for engineers who would rather write conversations, workflows and tools as files than drag nodes on a canvas. The company reports 750,000 agents built on the platform. The distinguishing technical piece is the Autonomous Engine, where an LLM decides how a conversation proceeds instead of following a hand-drawn decision tree — agents can call tools, run multi-step workflows, and take real actions like processing a refund or updating an account rather than routing every complication back to a human.
The billing model is the other thing that sets it apart. Botpress prices on conversations, not seats, and bundles LLM inference, embeddings and web search into the conversation allotment — so there is no separate OpenAI or Anthropic token bill to reconcile, and adding support staff does not raise the software cost. A conversation is any exchange with at least two end-user messages in the billing month, counted the same whether a bot or a human handled it. One important caveat for anyone who remembers the older Botpress: the open-source v12 edition and every self-hosted or locally installed version have been officially sunset. The v12 source is still on GitHub under AGPLv3 and existing v12 subscribers remain supported, but no new self-hosted deployments are possible. If you adopt Botpress in 2026, you are adopting Botpress Cloud.
Botpress fits teams that have outgrown a simple FAQ widget and need an agent that executes actions against real systems — but who do not want to build and host that agent themselves. The free tier at 100 conversations is enough to evaluate the Studio and ship a prototype, though the conversation-based pricing means real production volume lands on Plus ($150/mo for 250 conversations) or Team ($750/mo for 1,500 conversations) fairly quickly. The seat-free model makes it disproportionately attractive to organisations with large support teams and moderate ticket volume, and disproportionately expensive for the reverse.
Starting price: $150/mo · Free tier: yes · Model: freemium
Price history tracked from June 2026
| Plan | Price | Includes |
|---|---|---|
| Free | $0 | 100 conversations, no top-ups or overages · 3 seats and 3 AI agents · AI usage (LLM inference) included · Botpress Desk, Studio + ADK, Help Center · 1K table rows, 100MB vector, 100MB file storage · Community support via Discord |
| Plus | $150/mo | 250 conversations/mo included, billed annually · Extra packs of 100 conversations for $65 ($0.65/convo) · 3 seats, unlimited AI agents · WhatsApp channel and whitelabel webchat · Help Center with custom domain · 100K table rows, 1GB vector, 10GB file storage · Technical support via live chat |
| Team | $750/mo | 1,500 conversations/mo included, billed annually · Extra packs of 100 conversations for $50 ($0.50/convo) · Unlimited seats · Teams, routing rules and RBAC · Team analytics and real-time Studio collaboration · 500K table rows, 5GB vector, 50GB file storage |
| Enterprise | Custom | Custom conversation volume and storage · Voice channel (not available on lower tiers) · Dedicated support engineer · SLA and formal security review · Custom data retention and residency policies |
| Pros | Cons |
|---|---|
| No per-seat pricing — the Team plan gives unlimited seats, so support headcount does not raise the software bill | The open-source v12 self-hosted edition is officially sunset — no new downloads or deployments, so there is no way to run Botpress on your own infrastructure and the platform is full cloud lock-in |
| LLM inference, embeddings and web search are bundled into the conversation allotment; no separate provider token bill to reconcile | Auto-recharge on conversation overage cannot be turned off on paid plans; hitting the limit automatically buys packs of 100 rather than pausing |
| Genuinely free tier at 100 conversations with 3 seats, 3 agents and hard caps instead of surprise overages | The jump from Plus ($150/mo, 250 conversations) to Team ($750/mo, 1,500 conversations) is a 5x price step with no tier in between |
| Both a visual builder and a code-first TypeScript ADK, so non-developers and engineers can work in the same platform | Headline prices require annual billing to get the 20% discount; monthly billing costs more |
| SOC 2 certified, GDPR compliant, and penetration tested by KPMG — enough for regulated buyers | Voice channel is Enterprise-only, and storage beyond tier limits costs an extra $40/mo add-on |
| Emulator and testing conversations count against the paid conversation quota |
No-code platform for building and deploying AI support agents trained on your own business data
Open-source LLM app platform for building agentic workflows, RAG pipelines, and AI agents with a visual no-code builder
Open-source low-code platform to build LLM apps and AI agents with a drag-and-drop visual flow builder
Enterprise platform for branded customer-experience AI agents, billed on outcomes rather than seats or usage
Enterprise AI customer-service agents for chat, email, and voice, configured with natural-language workflows
Intercom's AI customer-service agent that resolves support tickets across chat, email, voice, and social — billed per resolution, not per seat
Yes, there is a permanent free tier at $0. It includes 100 conversations, 3 seats, 3 AI agents, Botpress Desk, Studio and the ADK, and a Help Center. LLM usage is included. The free tier is hard-capped with no top-ups or overages, and storage is limited to 1,000 table rows with 100MB of vector storage.
Plus is $150/mo with 250 conversations included; Team is $750/mo with 1,500 conversations and unlimited seats. Both prices are billed annually — annual billing carries a 20% discount, so monthly billing costs more. Enterprise is custom-priced. Extra conversations come in packs of 100 at $65 on Plus and $50 on Team.
Any exchange with at least two end-user messages in the billing month, whether handled by a bot, a human agent, or both. Emulator conversations count, spam does not. A conversation spanning two months counts in both, and each anonymous session counts separately.
No. The open-source v12 edition and all self-hosted or locally installed versions have been officially sunset — they are no longer available for purchase, download, or new deployments. The v12 source remains on GitHub under AGPLv3 for reference, and existing v12 subscribers are still supported, but all new users are directed to Botpress Cloud.
No. LLM inference, embeddings and web search are bundled into your conversation allotment, so there is no separate token bill. Each conversation you buy includes a proportional AI quota. If a workspace hits its AI limit before its conversation limit, a conversation pack is triggered.
No. Free and Plus include 3 seats; Team and Enterprise include unlimited seats. Pricing scales on conversation volume rather than headcount, which is the main structural difference from per-agent helpdesk tools.
Yes. The Agent Development Kit (ADK) is a TypeScript CLI framework — `adk init` scaffolds a project, `adk dev` runs it locally against a dev bot, and `adk deploy` ships it to Botpress Cloud. Conversations, workflows, tools, actions, tables, knowledge and evals each live in their own file.
Webchat, WhatsApp, Instagram, Messenger, Slack and Telegram among others. Webchat is on every tier, WhatsApp and whitelabel webchat start at Plus, and the voice channel is Enterprise-only.