# Inference Endpoints creation

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-endpoints-creation%2F64644%3E%5C
**Category:** Intermediate\
**Created:** [December 5, 2023, 11:22am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-endpoints-creation%2F64644 "2023-12-05T11:22:49Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![coreaiteam](https://avatars.discourse-cdn.com/v4/letter/c/7993a0/32.png) [@coreaiteam](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fcoreaiteam)%5C
**Post date:** [December 5, 2023, 11:22am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-endpoints-creation%2F64644%2F1 "2023-12-05T11:22:49Z")

</div>

I have finetuned a LLM model on my custom data using peft and lora on Autotrain Advance UI.  
Now how to create a inference endpoint after finetuning it?  
If anyone have idea about it or have done this before please share it.

---

<div class="post-metadata">

**Author:** ![iarbel](https://avatars.discourse-cdn.com/v4/letter/i/3da27b/32.png) [@iarbel](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fiarbel)%5C
**Post date:** [January 14, 2024, 11:05am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-endpoints-creation%2F64644%2F2 "2024-01-14T11:05:39Z")

</div>

+1 for that. text-generation-inference supports loading a PeftModel directly with the adapter, without the need to merge weights before. I’d expect the Inference Endpoints to support this option, however I couldn’t find a way to do it. Specifically for quantized models this is very important, as it’s not possible to merge LoRA weights into the base