Voiceflow named a 2026 Best Software Award winner by G2
Read now

Choosing an AI platform is less about the demo and more about what you'll be managing six months in. Botpress is one of the more capable builders on the market, and in 2025 it raised real money to push further. But "capable" and "right for your team" are two different questions.
Botpress Studio gives you a flexible, developer-friendly way to build agents, and its pay-as-you-go plan lowers the barrier to start. Here's an honest look at what Botpress does well, where it gets expensive, and the alternatives worth weighing before you commit.
Three things have changed since most Botpress reviews were written:
Botpress is a platform for building and deploying AI agents that handle tasks like customer service automation, onboarding, and lead capture. Botpress Studio is the build environment. The company also offers a hosted cloud, an integrations hub, and an API for wiring agents into your own apps and websites.
Like Voiceflow's visual builder, Botpress Studio uses a drag-and-drop canvas so your team can design conversations as flowcharts. You can import existing data such as knowledge bases, connect external tools, and layer LLM reasoning on top of structured flows. It leans developer-friendly: you'll get the most out of it if someone on the team is comfortable with light coding.
Teams use Botpress across a few common jobs:
Botpress ships a library of starter templates and integrations so you don't begin from a blank canvas, including playful ones like a text-adventure game master. By comparison, Voiceflow offers community-built templates across most common use cases, so you can usually find a starting point close to what you're building.
{{blue-cta}}
Botpress's API lets developers embed agents into existing workflows and pull data and analytics back out. If you want a flexible system that slots into your own stack rather than a closed product, this is one of Botpress's stronger points. It's also why Botpress appeals to engineering-led teams over no-code-only ones.
Botpress is led by CEO Sylvain Perron, who founded the company in 2017 with the goal of building chatbot solutions around client needs. After a $15M Series A in 2021, the company's June 2025 $25M Series B signaled a clear shift toward agent infrastructure. Investors include Framework Venture Partners, Inovia Capital, Decibel Partners, Deloitte Ventures, and HubSpot Ventures. Botpress counts brands like Kia, Electronic Arts, and Husqvarna among its referenced users.
Botpress moved to a usage-based model. You pick a base plan, and AI Spend (the cost of the underlying model calls) is billed on top of it. That's the number most first-time buyers miss. Here's the current structure:
Plan | What you get |
Pay-as-you-go | Free to start. 500 incoming messages/month, 1 seat, and a $5 monthly AI credit. Usage beyond the credit is billed as AI Spend. |
Plus | $89/month + AI Spend. 1 GB vector DB storage, human handoff for live support, AI conversation insights, and the option to remove Botpress branding. |
Team | $495/month + AI Spend. 3 seats, 2 GB vector DB storage, role-based access control, real-time Studio collaboration, and up to 50,000 incoming messages/month. |
Managed / Enterprise | Custom pricing. For larger deployments that need dedicated support, security reviews, and higher limits. Contact Botpress. |
The takeaway: the sticker price is the floor, not the ceiling. A Team plan at $495/month can cost meaningfully more once AI Spend is added, depending on conversation volume and which models you run. Model your real monthly usage before you commit.
Plenty of platforms build agents besides Botpress. A few worth comparing if you're shopping:
For a wider field, see our roundups of the best AI agent builders and the best AI chatbots.
Both platforms build capable agents. The real difference shows up in what you can run, which models you can use, and how much you can see once the agent is live.

Channels and models. Botpress is strong for chat and web, with voice available through integrations. Voiceflow ships native voice and phone day-one, and it's model-agnostic: you pick OpenAI, Anthropic, Google, Bedrock, or Groq per agent, or bring your own. If avoiding model lock-in matters to you, that's a real distinction.
Building and managing. Botpress gives you flexible flows plus autonomous reasoning. Voiceflow splits this into two primitives that compose: Workflows are deterministic SOPs for tasks that must go right every time (refunds, compliance checks), and Playbooks give the agent a goal and room to reason. A workflow can hand off to a playbook mid-conversation, and back again. That structure is what makes an agent adaptable without becoming unpredictable.
Seeing what the agent actually does. This is where most platforms thin out. Voiceflow includes a Knowledge Base, Evaluations, and an observability suite so you can trace every conversation, define what "good" looks like, and test changes in staging before they ship. Good agents, like good employees, need good management, and that's the layer Voiceflow is built around. It's also SOC 2 Type 2 compliant with PII masking for teams with compliance requirements.
Voiceflow's customers include Turo, StubHub International, Sanlam Studios, and Trilogy. If you want to see how the agent-management model works against your own use case, book a demo.
{{blue-cta}}
Botpress is used to build and deploy AI agents for customer service, onboarding, lead capture, and internal automation. Teams connect it to knowledge bases and external tools so agents can answer questions and take actions, not just chat.
Not entirely. Botpress has a free pay-as-you-go plan that includes a small monthly AI credit, but usage beyond that credit is billed as AI Spend. Paid plans (Plus and Team) add seats, storage, and higher message limits on top of that usage cost.
Botpress is a well-funded, credible platform. Its June 2025 $25M Series B and move toward agent infrastructure suggest it's investing for the long term. Whether it's the right fit depends on your team: it rewards developer-leaning teams more than no-code-only ones.
Botpress is SOC 2 compliant and, as an open-source platform, gives technical teams access to inspect and self-host parts of the stack. As with any AI tool, your safety also depends on how you configure guardrails, handle data, and review what the agent says in production.
Yes. Botpress's open-source roots let developers inspect the source and tailor the platform to their needs. That flexibility is part of its appeal for engineering teams, though it also means more of the setup and maintenance falls on you.