# Training LLM using local application

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-llm-using-local-application%2F84343%3E%5C
**Category:** Community Calls\
**Created:** [April 29, 2024, 1:05pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-llm-using-local-application%2F84343 "2024-04-29T13:05:16Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![sysley](https://avatars.discourse-cdn.com/v4/letter/s/bcef8e/32.png) [@sysley](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsysley)%5C
**Post date:** [April 29, 2024, 1:05pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-llm-using-local-application%2F84343%2F1 "2024-04-29T13:05:16Z")

</div>

Hi guys,  
as a AI newbie I’m looking to teach a LLM the way to use a local tool ; do i have to give the user manual to the LLM, to give the differrents possible actions, use tensorflow or pytorch?

Thanks for any help about the framework and AI tools to be used 🙂  
regards

---

<div class="post-metadata">

**Author:** ![swtb](https://avatars.discourse-cdn.com/v4/letter/s/8c91f0/32.png) [@swtb](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fswtb)%5C
**Post date:** [May 1, 2024, 10:06am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-llm-using-local-application%2F84343%2F3 "2024-05-01T10:06:41Z")

</div>

What is the local tool? What would the LLM prompts look like? What would you expect it to generate?

---

<div class="post-metadata">

**Author:** ![nielsr](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/nielsr/32/39617_2.png) [@nielsr](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fnielsr)%5C
**Post date:** [May 1, 2024, 1:38pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftraining-llm-using-local-application%2F84343%2F4 "2024-05-01T13:38:50Z")

</div>

One could fine-tune LLMs for so-called tool use. See the docs regarding how to do this with TRL: [Learning Tools (Experimental 🧪)](https://huggingface.co/docs/trl/en/learning_tools). It teaches an LLM how to use a calculator.

Note: this is pretty experimental.