> For the complete documentation index, see [llms.txt](https://aizen-corp.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://aizen-corp.gitbook.io/docs/managing-llm-workflows/llm-workflow.md).

# LLM Workflow

## Fine-Tuning and Serving LLMs

This diagram shows the workflow for fine-tuning and serving large language models (LLMs).

<div align="left"><figure><img src="/files/BFHpLNzDKT8hKaVgboWv" alt=""><figcaption><p>Workflow for LLMs</p></figcaption></figure></div>

Some of the steps in the workflow are optional. For example, you can directly serve a pretrained LLM without fine-tuning.

## Deploying Vector Stores and AI Agents

This diagram shows the workflow for serving LLMs or embeddings models for deploying vector stores for RAG applications or deploying AI agents.

<div align="left"><figure><img src="/files/W8wuRQbNdgExessePuCf" alt=""><figcaption><p>Workflow for Vector Stores and AI Agents</p></figcaption></figure></div>

Some of the steps in the workflow are optional. For example, you do not need to serve LLMs if the LLMs that you plan to use are from a non-Aizen vendor, such as OpenAI. Also, you are only required to configure a vector store if you plan to add a RAG Query tool to the AI agent.