You have engineers, and they want it
A team that will own the flows, the tests and the deploys gets more out of a platform than out of a partner. Botpress gives them a studio, an SDK and somewhere to put the logic.
Comparison
Olano vs Botpress
The short verdict: Botpress is a developer-first platform for building agents - a visual studio plus real code, an open-source lineage, a large integration library, and the option to self-host. It is a good tool, and if you have engineers who want to own the build, it is a reasonable place to put them. Olano is not a tool you build on; it is a system built, deployed and operated for you, with the parts a business needs around an agent - approval gates, an audit trail, spend caps and an isolated environment - shipped as product rather than as your integration work. The question is not which is more capable. It is who is doing the building, and who is on call at 2am when it breaks.
Side by side
The dimensions that decide the choice - not model benchmarks.
| Dimension | Botpress | Olano |
|---|---|---|
| Who does the building | You do. A visual studio plus code, with an open-source lineage and full control of the logic. | We do. Workflows are scoped, built, tested and launched by the Olano team, then kept running - your team reviews and approves. |
| Who operates it | You do: monitoring, breakages, model changes, and the on-call when a channel stops answering. | Managed operation is the product. Monitoring, upgrades and incident response sit with us, on an isolated deployment per customer. |
| Approvals & audit | Human handoff to a live agent is supported; anything more (who may approve what, and the record of it) is yours to design and build. | Trust levels 0-4, per-category approval gates, and an immutable audit trail - shipped, not built per project. |
| Channels | Web chat plus the usual messaging integrations, wired up per project. | WhatsApp, Telegram, email, Slack, SMS and 15+ messaging channels, plus inbound webhooks - configured as part of the engagement. |
| Multi-agent work | You compose it yourself from flows, tools and sub-bots. | An org chart of specialists that delegate to each other - 180+ archetypes, 19 starter teams - with each delegation gated and logged. |
| Pricing shape | Plans plus message volume: a free pay-as-you-go tier (500 incoming messages/month, one seat), Plus at $89/month, Team at $495/month (3 seats, up to 50,000 messages), Enterprise from around $2,000/month - with AI spend metered on top of the plan. | No per-seat pricing and no per-message meter for people talking to your agents. Olano Cloud from $10/month for a managed deployment; Studio engagements quoted as a fixed proposal, with a hard AI spend cap agreed before launch and BYOK at no extra cost. |
| Self-hosting & data | Self-hosting is available and is one of the platform's real selling points for data sovereignty. | Isolated deployment per customer as standard, custom AWS/GCP regions, or self-hosted on your own infrastructure - and we never hold your data in a shared tenant. |
| Models | Model-agnostic, with LLM usage billed as AI spend on top of the plan. | 18+ providers mixed per agent, with BYOK at no extra cost, inside a hard spend cap. |
Botpress's column reflects its publicly listed plans and documentation as summarised 3 September 2026; its pricing changed materially in May 2026, so check botpress.com/pricing for the current numbers before you decide on them.
Where Botpress is the better choice
Buying a managed system when you wanted a toolkit is an expensive mistake in the other direction. Botpress is the right answer more often than vendors like us admit.
Self-hosting is non-negotiable
If policy says the runtime lives on infrastructure you control and nothing else will do, self-hosting Botpress is a well-trodden path.
The scope is one well-understood bot
A support bot over a documented knowledge base, on your website, with handoff to a human. That is squarely what the platform is for, and you do not need an operating layer around it.
You want to start free and iterate
The pay-as-you-go tier costs nothing to try. Building the first version yourself is also the fastest way to learn what you actually need - including whether you want to keep operating it.
Where Olano is the better choice
Nearly every platform demo ends at launch. The interesting question is month four, when the model changed, an integration moved, and the person who built it is on another project.
Nobody on your team wants to operate it
The build is the small part. Olano's engagement covers design, integration, launch and the months afterwards - monitoring, upgrades and the on-call, on an isolated deployment.
It has to be governed, not just built
Approval gates before anything goes out under your name, trust levels 0-4, and an audit trail you can hand to a client or an auditor. On a build-it-yourself platform, that is your backlog.
The work is operational, not conversational
Triage an inbox, update the CRM, chase a follow-up, run a scheduled report, watch for an event and act on it. Agents that act across 450+ connector tools, not a conversation flow.
Spend must be predictable
A hard AI spend cap agreed before launch, a transparent usage meter, BYOK at no extra cost, and no per-message meter on the people talking to your agents.
On the pricing comparison, honestly
The two models are hard to compare on a spreadsheet and easy to compare on a bad month. Botpress bills a plan plus message volume plus AI spend, so a customer-facing bot that gets popular costs more precisely when it is working - and the free tier's 500 messages a month is a trial allowance, not a deployment. Olano does not meter the people talking to your agents at all: usage is AI spend on a transparent meter inside a cap you agree before launch. That is better for a public-facing front desk and worse if you want a near-zero bill for a bot nobody uses.
The comparison that actually matters is the one nobody puts on the pricing page: the engineering time. A platform licence is the smaller number next to a developer's month, and the operating cost does not stop after launch. If you have that engineer and they want the work, Botpress is cheaper in every sense. If you would be hiring or borrowing one to keep an agent alive, compare the whole line, not the subscription.
FAQ
Is Olano just Botpress with services on top?
No. Olano is its own runtime: persistent agents with memory, an org chart of specialists that delegate to each other, Cortex proposing improvements from the work already done, trust levels and approval gates, and an isolated deployment per customer. The services are how it is delivered, not what it is.
We already built something on Botpress. Is that wasted?
Not necessarily. Olano speaks MCP (200+ servers) and takes inbound webhooks, so an existing bot can stay where it is and hand work across rather than being rebuilt. We would rather compose with what works than charge you to replace it - see hosted AI agents vs DIY automation.
Can we self-host Olano?
Yes. The default is an isolated managed deployment per customer, with custom AWS/GCP regions available, and self-hosting on your own infrastructure is supported for buyers whose policy requires it. What we do not do is put your agents in a shared multi-tenant environment.
How do the numbers compare on a real workload?
It depends almost entirely on message volume and how much engineering you would be doing yourself, so any single figure would be marketing. What we can do is scope your actual workflow and put a fixed proposal against it - and tell you if a platform you operate yourself would be cheaper.
Are the Botpress figures here current?
They were summarised on 3 September 2026 from publicly listed plans, and Botpress restructured its pricing in May 2026 to meter AI spend alongside the plan. Treat the numbers as an order of magnitude and confirm them at botpress.com/pricing before deciding anything on them.
Related reading
Other comparisons, and the patterns behind them.
Hosted AI agents vs DIY automation
The build-vs-buy question in full, including when DIY wins.
Olano vs Voiceflow
The design-led conversational platform, compared on the same axes.
AI agent platform vs chatbot
Why "chatbot" is the wrong mental model for agents that act.
How to choose an AI agent platform
Channels, approvals, audit, spend caps, BYOK, isolation - the checklist.
Checked 3 September 2026 against the vendor's own published material: Botpress pricing · Botpress pricing update, May 2026. Comparisons go out of date - if we have described your product wrongly, write to support@olano.ai and we will correct this page.
Not sure whether you want a platform or a partner?
Bring one workflow. We will map it, tell you plainly whether your own team should build it on a platform like Botpress, and quote only if the answer is no.