Amongst the platforms that enable users to build LLM-powered bots, with or without the need for coding knowledge, Botpress and Voiceflow seem to be some of the most discussed and beloved.
In this article, we ask ourselves which of Voiceflow and Botpress is the best AI agent builder, discussing:
- Where did Botpress and Voiceflow, respectively, come from and where are these platforms today
- What do Botpress and Voiceflow have in common & what standout features does each have
- When does it make more sense to use Botpress & when is Voiceflow more recommended
- Future plans & roadmaps: what can we expect from Botpress and Voiceflow?
Botpress vs. Voiceflow: Where they came from & What are they today
Botpress: From dev-friendly CAI platform to the GPT pivot
Born in 2015 from the combination of CEO Sylvain Perron’s realization of a bot-platform-shaped gap in the market, and Decibel and Real Ventures funding, the Botpress company was created with the goal of filling just that hole.
One year later, bots were starting to become really popular. Thanks to their decision to make Botpress open-source, the platform quickly became a Github favorite. Combining an NLU engine, a conversation design studio and flexible integrations, the application provides developers with everything they need to build, deploy, and manage chatbots on their own.
When the first Fortune 500 came knocking on their door, Botpress was probably pretty chuffed with realizing that there was a way for them to monetize, too. Throughout the years, they’ve been able to attract prestigious customers, such as Siemens, Asus, Kia, and EA, and, in 2021, to nab a $15 million Series A from Decibel and Inovia Capital.
Still, a more defining change would come for the platform.
In March 2023, after two months of testing with over 20,000 beta users, Botpress announced a pivot into Gen AI, becoming “the industry's first visual builder tool for creating chatbots that can execute sophisticated workflows on top of ChatGPT in a matter of minutes.”
Voiceflow: Road to becoming the “Figma of Conversational AI”
Braden Ream, Tyler Han, Michael Hood and Andrew Lawrence were college friends, with a passion for building Alexa apps together. The more agents they built, the more frustrated they would get with the platforms available at the time, motivating them to launch Voiceflow, in 2019.
Voiceflow was born as a tool to allow conversation designers to quickly and intuitively build conversational assistants, giving them a way to more fluidly collaborate with their team.
Voiceflow quickly expanded beyond the Alexa ecosystem. In time, the platform has become known in the industry as designers' favorite, and, with its continuous updates and releases, Voiceflow proves to its customers time and again that they can be trusted to be on top of emerging trends and needs.
As CEO Braden Ream highlights, in the last two years, they’ve “successfully built the Figma for conversational interfaces”, used by 130,000 people worldwide, including various Fortune 500s, like Amazon, BMW and US Bank. Voiceflow has seen an exponential growth in 2023, the year they’ve launched lots of new Gen AI-powered features, like the LLM-powered knowledge base.
In August 2023, Voiceflow raised a new funding round of $15 million USD, led by Venture Partners, to support the future ambitions of the company (more on that later).
Botpress vs. Voiceflow: The best AI agent builder
What do Botpress and Voiceflow have in common
Voiceflow and Botpress are two of the most popular AI agent platforms on the market. As we will discuss in the next section, each of the two has its unique strengths, but there are some characteristics that you can expect to find in both.
Some of the most popular features that Voiceflow and Botpress have in common are:
- The low-code drag-and-drop canvas, where designers can create flows with logic, using different input and output modalities, account for different paths and test their WIP projects. Both platforms are accessible to users without coding knowledge, but they offer ways to make it easy to cater to more advanced use cases that require some code (see Codex-powered transitions and actions for Botpress, and Custom Code block for Voiceflow)
- Enabling the combination of generative and traditional bot building, allowing users to decide when it makes more sense to create a rule-based component, and leverage the power of LLMs sensibly.
- The LLM-powered knowledge base allows the bot to respond about domain-specific topics, relying on uploaded URLs and documents.
- The ability to collaborate with colleagues and clients on the same project.
- Dashboards & analytics, to help users see how their bot is performing and identify opportunities for improvement.
- The ability to quickly create a shareable prototype, so that internal and external testers can give feedback on a POC.
What makes Botpress stand out
While Botpress and Voiceflow have some similarities with regards to the bot-building features they offer, there are some differences that make each stand out on their own. These are some of Botpress’s standout capabilities:
- Lots of direct interactions with many popular business apps, including Gmail, GitHub, Salesforce, Slack, Twitter, Zendesk, Zoom, and a Zapier connector.
- Multi-language support, which allows users to build in one language and directly obtain automatic translations for over 100 others.
- Ability to use videos and files as output (i.e. in the bot’s message). This isn’t often possible in no-code platforms, but it might come in handy, depending on the application.
- Codex-powered transitions and actions, allowing users with no developing skills to describe their goal in natural language and have it transformed into working code.
What makes Voiceflow standout
On the other hand, Voiceflow also has some standout features that distinguish it from Botpress. These include:
- Its highly accurate and hyper-fast NLU engine, which seems to outperform various industry leaders.
- For AI-generated responses, users can choose between different LLM models from OpenAI and Anthropic at various degrees of accuracy and price level.
- Multimodality, as Voiceflow supports chatbot, voicebots, IVRs, and Alexa-related projects.
- The drag-and-drop studio features tools to design more advanced use cases and features without the need for coding knowledge. For example, API steps create a form to quickly set up an external call.
- Lots of templates for popular and advanced use cases, as well as tutorials on how to deploy assistants on popular channels.
When to use Botpress vs. Voiceflow
Comparing Botpress and Voiceflow, it’s become clear why these are two of the most popular AI agent building platforms out there. If you’re looking for a tool to build an AI assistant, combining NLU with LLM-powered capabilities, these are excellent options. Still, each one can be better suited for different situations.
You might prefer Botpress, for example, if you’re looking to build a bot that’s supported in multiple languages and the applications you’re envisioning are not too advanced.
On the other hand, if you have more specific requirements and you’re good with a little bit of coding, Voiceflow will offer a more flexible and powerful alternative. Voiceflow also appears to be the better alternative for varied teams composed of figures with different backgrounds and goals, as it allows conversation designers, product managers and developers to communicate more effectively.
Botpress vs. Voiceflow: Roadmap & Future plans
One of the aspects we always recommend paying attention to, when evaluating any conversational AI tool or platform, is to spend some time researching what their future plans and their related roadmap look like.
At the time of writing, Botpress has revealed their plans to release new LLM-enabled features, like a sentiment analysis dashboard, and a Natural Language Generation tool to give designers proactive content suggestions and speed up build time.
Meanwhile, Voiceflow has announced they are “investing further in Voiceflow’s end-to-end agent building and hosting capabilities”, to be able to support conversational AI teams not only in design and prototyping phases but to become an allround, end-to-end Conversational AI platform.
Inspired to explore what conversational AI can do for your company? We can help.
And if you’re still hungry for knowledge, follow us on LinkedIn for weekly updates on the world of conversational AI, or check out our article on what we like to call the Conversational AI Metrics Game, to learn all about how to measure chatbots.