---
title: "Tray.ai alternative: FlowRunner vs Tray.ai"
description: "FlowRunner vs Tray.ai: a permanent free plan, published prices, one execution per run, and waits of up to a year inside one run. Where each is stronger, and how to choose."
url: https://flowrunner.ai/alternatives/tray-io
date_published: 2026-11-02
date_modified: 2026-11-03T09:28:56-08:00
---

[Alternatives](https://flowrunner.ai/alternatives) / Tray.ai

![Tray.ai logo](https://flowrunner.ai/assets/alternatives/tray-io.svg)

# Tray.ai alternative: FlowRunner vs Tray.ai

FlowRunner vs Tray.ai: a permanent free plan, published prices, one execution per run, and waits of up to a year inside one run. Where each is stronger, and how to choose.

2026-11-02 Updated November 3, 2026 11 min

## Use FlowRunner when

-   You want a free plan that keeps running after the trial ends
-   A run has to wait days or months for a person or a date, and stay one run
-   You'd rather count whole runs than every operation on your data
-   AI agents and approvals should come with every plan, not as add-ons
-   You want to see the price on the pricing page

## Use Tray.ai when

-   Your workflows run Python steps and you want that code inside the workflow
-   You want 30 days and 1 million tasks to evaluate before choosing a plan
-   Audit trails need to come with the entry plan you're quoted

**TL;DR**

-   Start with what these two share: connectors to the apps you run, triggers, branches and loops, approval steps, and AI agents. FlowRunner does all of that, and it has a permanent Free plan and prices you can read on its pricing page.
-   FlowRunner counts one execution per complete run. Tray.ai counts a task for each operation it performs on your data.
-   A FlowRunner run waits up to 30 days on Free, Starter, and Growth, and up to a year on Professional and above. Tray.ai’s Delay step tops out at 10 minutes, and for longer waits Tray recommends saving state and running a second workflow on a daily schedule.
-   Tray.ai is stronger on the size of its trial, Python steps inside a workflow, and audit trails listed on every plan.

## Who this comparison is for

You connect business apps and automate the work between them, and more of that work now has an AI agent in it. Some steps need a person to look before anything happens, and some runs have to wait for days: a contract signature, a customer reply, a renewal date. You might be on Tray.ai today, or weighing it against FlowRunner before you commit.

The overlap is large. Tray workflows connect apps, react to webhooks and schedules, branch, loop, and call other workflows. FlowRunner flows do the same. The difference shows up in three places: what you pay for, how long a single run can wait, and whether agents and approvals come with the plan. This page goes through each.

## What FlowRunner is good at

FlowRunner is built around agents that know when to stop and ask a person. An AI agent in a flow can call one of your own flows as a tool. That flow sends the question by email, Slack, WhatsApp, phone, or a [Form](https://docs.flowrunner.ai/platform/forms.html), holds the run until the person answers, and hands the answer back. The agent picks up where it stopped. This is [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) the agent decides on at runtime.

FlowRunner is strong on:

-   **One run, one execution.** An execution is one complete run, however many steps it has and however long it waits.
-   **Waiting inside the run.** A run can wait up to 30 days on Free, Starter, and Growth, and up to a year on Professional and above, with its data and context intact.
-   **Prices you can read.** Free is $0 for 100 executions a month. Starter is $5 (300) or $15 (3,000). Growth is $45 (12,000), $89 (30,000), or $149 (60,000). Professional is $299 (75,000), Business is $999 (250,000), and Enterprise is a custom quote.
-   **Agents and approvals on every plan.** Every integration, AI agents with [BYOK](https://flowrunner.ai/concepts/byok), human-in-the-loop, and unlimited users and workflows come with every plan, including Free.
-   **Compliance at listed prices.** Audit trails and RBAC start at Professional ($299). SSO/SAML, SLA tracking, and HIPAA with a BAA come with Business ($999).
-   **A deep, agent-ready catalog.** 2,100+ integrations and 56,000+ callable actions (as of September 2026), roughly 26 actions per integration, each with the typed parameters and descriptions an agent needs to call it as a tool.

## What Tray.ai is good at

Tray.ai describes itself as “one unified platform for data and AI,” for teams that deploy, run, and govern integrations, automations, agents, MCP, and apps. It sells to companies where IT and business teams build on the same platform.

Tray.ai is built for:

-   **Promoting work between environments.** Projects, project versions, and import and export move workflows, data table structure, and API endpoint definitions between workspaces.

## Where FlowRunner is stronger

-   **Waits that stay in one run.** Tray.ai’s Delay step has a maximum of 10 minutes. For anything longer, Tray’s docs say not to use it, and suggest saving the execution details and target date to data storage, then running a separate workflow on a daily schedule that compares dates and sends the data on. In FlowRunner the run itself waits, up to 30 days on Free, Starter, and Growth, and up to a year on Professional and above. One flow, one run record, and the context stays with it.
-   **A meter that doesn’t climb with steps.** Tray counts a task for each operation it performs on a piece of your data, plus Helix Compute Units, and settles overage at a true-up. FlowRunner counts one execution per complete run. Adding a step, or looping over a longer list, doesn’t change what a run costs.
-   **A free plan, and prices on the page.** Tray.ai has no free plan, and its Pro, Team, and Enterprise plans are “sized to your needs” with no published prices. FlowRunner’s Free plan is permanent, and every paid plan has a price on the [pricing page](https://flowrunner.ai/pricing).
-   **Agents and approvals included.** Tray sells its Merlin Agent Builder as an add-on priced on a demo, and Agent Gateway for MCP as another add-on. FlowRunner’s AI agents come with every plan, including Free, and run on your own provider keys with no markup on inference.
-   **Human-in-the-loop as an agent tool.** Tray’s approval steps route to Slack, email, or a custom form and resume when approved. FlowRunner’s approvals reach people by email, Slack, WhatsApp, phone, or a Form, on every plan, and an AI agent can call a review flow as a tool when it decides it needs a person. The run waits for the answer, up to the plan’s limit, and carries on.
-   **Compliance at listed prices.** Tray includes SSO and HIPAA in Enterprise and sells them as add-ons on Team, with the BAA through sales. FlowRunner’s RBAC comes with Professional at $299, and SSO/SAML and a BAA come with Business at $999, both at listed prices.
-   **The catalog.** FlowRunner has 2,100+ integrations to Tray’s 700+ connectors, with every integration on every plan, and every one of the 56,000+ actions carries the metadata an agent needs to call it as a tool.

## Where Tray.ai is stronger

-   **A bigger trial.** Tray’s trial runs 30 days with 1 million tasks, no card, and no sales call. FlowRunner’s trial is 14 days of Professional. When Tray’s trial ends there is no free plan to fall back to, while a FlowRunner workspace moves to Free and its flows keep running.
-   **Python steps.** Tray’s Python script connector adds Python functions to a workflow. FlowRunner’s code steps are JavaScript ([Custom Cloud Code](https://docs.flowrunner.ai/reference/custom-cloud-code.html), with `fetch` for network calls), and npm packages run in [custom extensions](https://docs.flowrunner.ai/extend/index.html). Python code stays where it runs today, as a service a flow calls with an HTTP Request block.
-   **Audit trails on every plan.** Tray lists audit trails on all of its plans. FlowRunner’s start at Professional. Tray’s entry plan has no published price, so the comparison is a quote against $299 on a page.

## Starting free

Tray.ai has no free plan. Its trial gives 30 days and 1 million tasks on Tray iPaaS, with no card and no sales call. After that, every plan is quoted.

FlowRunner has a permanent Free plan: 100 executions a month, every integration, AI agents on your own keys, human-in-the-loop, unlimited users and workflows, and 24 hours of log history. Every new account also gets 14 days of Professional with no card, so the trial covers audit trails and RBAC. When it ends, the workspace moves to Free and its flows keep running. Nothing is paused or deleted.

## Feature comparison

| Dimension | FlowRunner | Tray.ai |
| --- | --- | --- |
| Published prices | Yes, from $0 | No; Pro, Team, Enterprise sized by quote |
| Free plan | Yes, 100 executions a month, permanent, no card | No |
| Trial | 14 days of Professional, no card, then Free | 30 days and 1 million tasks, no card |
| Billing unit | Execution (one complete run) | Task (one operation on a piece of data), plus Helix Compute Units |
| Users | Unlimited on every plan | No per-user fees |
| Integrations | 2,100+ integrations, 56,000+ actions, every plan | 700+ connectors |
| Timed delay | Wait block; the run waits up to its plan’s limit | Delay step, up to 10 minutes |
| Longest wait in one run | 30 days (Free, Starter, Growth); 1 year (Professional and up) | Not documented; Tray recommends stored state and a daily scheduled workflow past 10 minutes |
| AI agents | Every plan, BYOK with no markup | Merlin Agent Builder, a paid add-on |
| MCP | Agents use tools from registered MCP servers; FlowRunner Prompt to Flow lets AI tools build and run flows | Headless MCP in all plans; Agent Gateway add-on |
| Human-in-the-loop | Every plan; an agent can call a review flow as a tool | Approval steps; agent approvals in Merlin (add-on) |
| Where people answer | Email, Slack, WhatsApp, phone, or a Form | Slack, email, or a custom form |
| Code steps | JavaScript (Custom Cloud Code); npm in custom extensions | JavaScript (Script) and Python (Python script) |
| Run history visible in-app | 24 hours (Free, Starter) to unlimited (Enterprise) | Logs kept 7 days on Pro and Team (expandable on Team) |
| Audit trails | Professional ($299) and up | Listed on all plans |
| SSO/SAML | Business ($999) and up | Enterprise; add-on on Team |
| HIPAA / BAA | BAA on Business ($999) and up | Enterprise, add-on on Team; BAA via sales |
| Where it runs | FlowRunner cloud; self-hosted on Enterprise | Tray cloud; region-specific hosting on Enterprise |
| Build from an AI chat | FlowRunner Prompt to Flow (Claude, ChatGPT, Codex, Cursor) | Tray Headless (Claude Code, Codex) |

## Decision framework

**Pick FlowRunner if:**

-   Runs have to wait hours, days, or months for a person or a date, and you want each one to stay a single run.
-   You want a free plan that keeps going after the trial, and prices you can read before you talk to sales.
-   You’d rather count whole runs than every operation on your data.
-   Your agents need to stop and ask a person, on every plan, by email, Slack, WhatsApp, phone, or a Form.
-   You want RBAC at $299 and SSO and a BAA at $999, at listed prices.

**Pick Tray.ai if:**

-   Your workflows depend on Python steps that should stay inside the workflow.
-   You want 30 days and 1 million tasks to evaluate before you choose a plan.
-   You need audit trails on the entry plan and are buying through a quote either way.

## Ready to move?

Tray.ai exports each workflow, or a whole project version, as JSON, and FlowRunner Prompt to Flow can turn each workflow into a flow from one chat. The [migration guide](https://flowrunner.ai/alternatives/tray-io/migrate) covers the export, the request to paste, and what to finish by hand. If you’d like to talk it through first, [contact us](https://flowrunner.ai/contact).

## Frequently asked questions

### Is Tray.ai the same as Tray.io?

Yes. The company now calls itself Tray.ai and its site is tray.ai. Many older articles still call it Tray.io.

### Does Tray.ai have a free plan?

No. Tray.ai offers a trial of 30 days and 1 million tasks with no card and no sales call, and after that its Pro, Team, and Enterprise plans are sized by quote. FlowRunner has a permanent Free plan with 100 executions a month, every integration, AI agents on your own keys, human-in-the-loop, and unlimited users. New accounts also get 14 days of Professional with no card, then move to Free with their flows still running.

### What counts as a task in Tray.ai?

Tray.ai defines a task as one operation Tray performs on a piece of your data, so a workflow with several steps uses several tasks each time it runs. FlowRunner counts one execution per complete run, however many steps it has and however long it waits.

### How long can a Tray.ai workflow wait?

Tray.ai's Delay step has a maximum of 10 minutes. For longer waits, Tray's docs say not to use it and suggest saving the details and the target date to data storage, then running a separate workflow on a daily schedule that picks the data up when the date arrives. In FlowRunner the run itself waits, up to 30 days on Free, Starter, and Growth, and up to a year on Professional and above, with its data and context intact.

### Does Tray.ai support human-in-the-loop approvals?

Yes. Tray.ai's process automation adds approval steps that route to Slack, email, or a custom form and resume when approved, and its Merlin Agent Builder, a paid add-on, adds approvals for sensitive agent actions. FlowRunner includes human-in-the-loop on every plan, including Free. An AI agent can call a review flow as a tool, the person answers by email, Slack, WhatsApp, phone, or a Form, and the agent carries on with the answer.

### Are AI agents included in Tray.ai plans?

Tray.ai sells its Merlin Agent Builder as an add-on, priced on a demo, and Agent Gateway for MCP as another add-on. Headless MCP is included in all plans. FlowRunner's AI agents come with every plan, including Free, and run on your own provider keys with no markup on inference.

### How do I move from Tray.ai to FlowRunner?

Export each workflow from the Tray builder as JSON, or export a whole project version. Then give one workflow at a time to FlowRunner Prompt to Flow in Claude, ChatGPT, Codex, or Cursor, and it builds the same workflow as a FlowRunner flow. Authentications don't travel, so you connect your accounts and test before switching over. The [step-by-step guide](https://flowrunner.ai/alternatives/tray-io/migrate) walks through it.

## Keep reading

-   [How to move from Tray.ai to FlowRunner](https://flowrunner.ai/alternatives/tray-io/migrate)
-   [FlowRunner vs Workato](https://flowrunner.ai/alternatives/workato)
-   [FlowRunner vs MuleSoft](https://flowrunner.ai/alternatives/mulesoft)
-   [FlowRunner vs Zapier](https://flowrunner.ai/alternatives/zapier)

[All alternatives](https://flowrunner.ai/alternatives)

By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy)

> _I like that aspect of the human in the loop when it's needed. That's ultimately what we need to get to._

VP Operations, CPG brand

---
Markdown version of https://flowrunner.ai/alternatives/tray-io. Site index: https://flowrunner.ai/llms.txt
