# TinyML integration

**URL:** <https://forum.rasa.com/t/tinyml-integration/38617>\
**Category:** Rasa Open Source\
**Created:** [January 4, 2021, 7:06am UTC](https://forum.rasa.com/t/tinyml-integration/38617 "2021-01-04T07:06:03Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![arkaprabha-majumdar](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/arkaprabha-majumdar/32/12446_2.png) [@arkaprabha-majumdar](https://forum.rasa.com/u/arkaprabha-majumdar)\
**Post date:** [January 4, 2021, 7:06am UTC](https://forum.rasa.com/t/tinyml-integration/38617/1 "2021-01-04T07:06:03Z")

</div>

Hey, does anyone know how I can convert a rasa chatbot to TinyML?

---

<div class="post-metadata">

**Author:** ![koaning](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/koaning/32/6626_2.png) [@koaning](https://forum.rasa.com/u/koaning)\
**Post date:** [January 4, 2021, 8:22am UTC](https://forum.rasa.com/t/tinyml-integration/38617/2 "2021-01-04T08:22:13Z")

</div>

> [@arkaprabha-majumdar](#):
>
> TinyML

I’m not sure if I’ve heard of TinyML. Do you have a link?

In general, though, a Rasa pipeline can contain pre-trained word embeddings as well as scikit-learn components. So the odds of being able to make it “tiny” is small.

---

<div class="post-metadata">

**Author:** ![arkaprabha-majumdar](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/arkaprabha-majumdar/32/12446_2.png) [@arkaprabha-majumdar](https://forum.rasa.com/u/arkaprabha-majumdar)\
**Post date:** [January 4, 2021, 11:26am UTC](https://forum.rasa.com/t/tinyml-integration/38617/3 "2021-01-04T11:26:03Z")

</div>

I’m new to tinyML myself… apparently its an optimized form of ML that can be run on much smaller devices like microprocessors

---

<div class="post-metadata">

**Author:** ![koaning](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/koaning/32/6626_2.png) [@koaning](https://forum.rasa.com/u/koaning)\
**Post date:** [January 4, 2021, 1:16pm UTC](https://forum.rasa.com/t/tinyml-integration/38617/4 "2021-01-04T13:16:12Z")

</div>

I see. Rasa is meant to be run on a server for now. We don’t support microprocessors.

There have been community projects to run it on a raspberry pi though [Run Rasa Docker image on your Raspberry Pi](http://forum.rasa.com/t/run-rasa-docker-image-on-your-raspberry-pi/30985).