# Unable to train rasa through endpoint

**URL:** <https://forum.rasa.com/t/unable-to-train-rasa-through-endpoint/42883>\
**Category:** Rasa Open Source\
**Created:** [April 26, 2021, 7:24am UTC](https://forum.rasa.com/t/unable-to-train-rasa-through-endpoint/42883 "2021-04-26T07:24:24Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![nirmal\_mahen](https://avatars.discourse-cdn.com/v4/letter/n/65b543/32.png) [@nirmal\_mahen](https://forum.rasa.com/u/nirmal_mahen)\
**Post date:** [April 26, 2021, 7:24am UTC](https://forum.rasa.com/t/unable-to-train-rasa-through-endpoint/42883/1 "2021-04-26T07:24:24Z")

</div>

So I have pulled the latest image from the docker and running it locally exposing it’s port with a docker-compose file of my own.

When I hit the train URL i.e,

> [http://localhost:5005/model/train?save\_to\_default\_model\_directory=true&force\_training=false&callback\_url=](http://localhost:5005/model/train?save_to_default_model_directory=true&force_training=false&callback_url=)

I get a **404 Not found** reponse…

But when i run the rasa server through terminal… using the following cmd

> rasa run -m models --enable-api --cors “\*” --debug

and hit the train api, The training sequence is established, Does any one know why?
