# Rasa CALM: Could not import huggingface\_hub python package

**URL:** <https://forum.rasa.com/t/rasa-calm-could-not-import-huggingface-hub-python-package/61844>\
**Category:** Rasa CALM\
**Tags:** installation, rasa\
**Created:** [May 5, 2024, 5:39pm UTC](https://forum.rasa.com/t/rasa-calm-could-not-import-huggingface-hub-python-package/61844 "2024-05-05T17:39:58Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![steveplexagon](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/steveplexagon/32/24144_2.png) [@steveplexagon](https://forum.rasa.com/u/steveplexagon)\
**Post date:** [May 5, 2024, 5:39pm UTC](https://forum.rasa.com/t/rasa-calm-could-not-import-huggingface-hub-python-package/61844/1 "2024-05-05T17:39:58Z")

</div>

When attempting to run a model from Huggingface, getting an error about the H.F. package even though I have loaded it into the environment with pip as per the error msg:

_“Please install it with `pip install huggingface_hub`. (type=value\_error)”_

here’s the confi.yml

```auto
recipe: default.v1
language: en
pipeline:
- name: KeywordIntentClassifier
- name: NLUCommandAdapter
- name: LLMCommandGenerator
  llm:
    type: "huggingface_hub"
    repo_id: "mistralai/Mistral-7B-v0.1"
    task: "text-generation"

policies:
- name: FlowPolicy
- name: EnterpriseSearchPolicy
- name: IntentlessPolicy

```

I have tried many different versions of this pipeline with no luck so far. Please advise.

P.S. I was able to successfully pipeline a custom fine-tuned gpt model which is very cool !

---

<div class="post-metadata">

**Author:** ![steveplexagon](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/steveplexagon/32/24144_2.png) [@steveplexagon](https://forum.rasa.com/u/steveplexagon)\
**Post date:** [May 7, 2024, 7:17pm UTC](https://forum.rasa.com/t/rasa-calm-could-not-import-huggingface-hub-python-package/61844/2 "2024-05-07T19:17:21Z")

</div>

update on this issue.

I had been hasty when I ran through demo install and neglected to update poetry …

- pip install huggingface\_hub

- poetry add huggingface\_hub

that made import hugginface\_hub work…

now I’m on to the challenge of getting the pipeline to work with a huggingface AWS endpoint…

for models like meta-llama/Meta-Llama-3-8B … a URL is given ( paid versions )

**Any thoughts about how to configure an LLM endpoint via AWS url, please share… !**

Thanks

---

<div class="post-metadata">

**Author:** ![Sanjukta.bs](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/sanjukta.bs/32/24453_2.png) [@Sanjukta.bs](https://forum.rasa.com/u/Sanjukta.bs)\
**Post date:** [August 3, 2024, 11:12am UTC](https://forum.rasa.com/t/rasa-calm-could-not-import-huggingface-hub-python-package/61844/3 "2024-08-03T11:12:56Z")

</div>

Can you share how you worked with hugging face? Because I want to use rasa with a local llm. Any helps will be highly appreciated.