Connect it. Deploy it. Own it.

ToRun doesn't live in a box. Push AI-generated content to the tools your audience already uses, let outside systems kick off your automations, and - if your industry demands it - run the entire platform inside your own walls, against your own models. No connector fees. No surprise line items. Every external call is metered at provider cost plus a margin we publish openly.

Build a workflow · See deployment options

Send your output where it belongs

Any workflow can end in a Publish step. Today that means a generic webhook - Slack, Discord, or any HTTP endpoint you control - or saving the result as a Markdown artifact in your knowledge base for reuse. Build a research brief, generate a weekly digest, draft a social post; then deliver it automatically, on schedule or the moment a run completes.

Because publishing is just another step in the graph, you can fan a single result out to several destinations at once - post to Slack, archive to your knowledge base, and fire a webhook to your own backend - all from one run, each delivery recorded.

OAuth-based social connectors (LinkedIn, Substack, X, Bluesky, Mastodon) are rolling out next, with their tokens vaulted and auto-refreshed so you set them once and forget them.

Let the outside world trigger your work

Workflows aren't only something you start. An inbound webhook lets GitHub, a CRM, a form tool, or your own backend fire a ToRun automation the instant something happens. Combine that with the HTTP Request node and OAuth token handling, and your workflows can both pull data from external APIs and react to events from them - no glue code to maintain.

Every connector you bind is defined explicitly, its credentials encrypted, and every call it makes is logged - so an integration is something you can audit, not a black box wired into your account.

Run it on your own ground

Some teams can't put data in anyone else's cloud. For them, ToRun ships as an on-prem deployment - the full platform inside your private network, brought up with a single command and pointed at your own local models (Ollama, vLLM, LM Studio) so nothing ever leaves your perimeter. It's billed per seat per year, with maintenance bundled in.

Banks and regulated industries can go further with a source-code license: a perpetual snapshot of the codebase for unlimited internal seats, paired with an annual Platform Evolution & Expert Knowledge subscription that keeps new models, providers, and security hardening flowing - plus architect advisory hours and priority SLA. A signed auto-update channel keeps you current without a migration project, and a code-escrow option is available for extra peace of mind.

On-prem is a first-class product, not an afterthought - the full story is on the On-Prem & Self-Hosted page.

Pricing you can actually read

This is the part most platforms hide. With ToRun, every integration call is itemized in a billing record - the external API, the tokens or compute it used, the rate applied. There is no "integration tier" tax bolted on top. Usage is billed at provider cost plus the same transparent margin you see everywhere else on the platform. What you spend is what it cost, plus a number you can point to.

What you can connect

• Publish to Slack & any webhook bi-send-check - end a workflow with a Publish step and deliver the result to Slack, Discord, or any HTTP endpoint - on a schedule or the moment a run completes, to several destinations at once.
• Save as reusable artifacts bi-file-earmark-richtext - drop any output into your knowledge base as a Markdown artifact, ready to feed the next workflow, chat, or research run.
• Inbound webhook triggers bi-box-arrow-in-down-right - let GitHub, your CRM, a form, or your own backend kick off a ToRun automation the instant an event fires.
• Call external APIs with OAuth bi-globe - the HTTP Request node pulls data from any API, with vaulted OAuth tokens refreshed automatically and every call logged.
• On-prem on your own models bi-hdd-network - run the full platform in your private network against your own local LLMs, or license the source for unlimited internal seats, with signed updates and architect advisory.
• Every call itemized - no connector tax bi-receipt - each integration call shows up in a billing record at provider cost plus a published margin. No hidden integration fees.