# Running multiple prodigy instances

**URL:** <https://support.prodi.gy/t/running-multiple-prodigy-instances/3093>\
**Category:** Uncategorized\
**Tags:** usage, solved, server\
**Created:** [June 27, 2020, 12:29pm UTC](https://support.prodi.gy/t/running-multiple-prodigy-instances/3093 "2020-06-27T12:29:33Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![mihaivinaga](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/mihaivinaga/32/1180_2.png) [@mihaivinaga](https://support.prodi.gy/u/mihaivinaga)\
**Post date:** [June 27, 2020, 12:29pm UTC](https://support.prodi.gy/t/running-multiple-prodigy-instances/3093/1 "2020-06-27T12:29:33Z")

</div>

Hello,

I have the following problem. I have two annotators and I want to start multiple prodigy instances because I want them to work on different datasets. I have searched the documentation and also this forum, but couldn't find anything regrading this. Is it possible to do such a thing?

---

<div class="post-metadata">

**Author:** ![ronaldoviber](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ronaldoviber/32/795_2.png) [@ronaldoviber](https://support.prodi.gy/u/ronaldoviber)\
**Post date:** [June 27, 2020, 4:01pm UTC](https://support.prodi.gy/t/running-multiple-prodigy-instances/3093/2 "2020-06-27T16:01:36Z")

</div>

yes, its possible:

> **[COVID-19-Text-Mining/prodigy-multitask](https://github.com/COVID-19-Text-Mining/prodigy-multitask)**
>
> Multitasking Docker image for Prodigy. Contribute to COVID-19-Text-Mining/prodigy-multitask development by creating an account on GitHub.

good luck!

---

<div class="post-metadata">

**Author:** ![ines](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ines/32/3_2.png) [@ines](https://support.prodi.gy/u/ines)\
**Post date:** [June 27, 2020, 6:19pm UTC](https://support.prodi.gy/t/running-multiple-prodigy-instances/3093/3 "2020-06-27T18:19:03Z")

</div>

> [@mihaivinaga](#):
>
> I have two annotators and I want to start multiple prodigy instances because I want them to work on different datasets.

Sure, the solution is probably much simpler than you think: When you start Prodigy, it just spins up a web server on the given host and port – so if you want to start multiple instances, you can start them on different ports, by setting the `PRODIGY_PORT` env variable. That's the simplest approach.

You can make it more sophisticated and use a reverse proxy to assign different subdomains, or even have a service that starts and stops instances automatically. The solution @ronaldoviber linked is an example of this – but none of that is required.