> Markdown version of [/events/world-congress-2024/sessions/170-efficient-large](https://www.wearedevelopers.com/events/world-congress-2024/sessions/170-efficient-large). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Efficient Large Language Model Customization with NVIDIA NeMo Framework.


- **Date:** Thursday, Jul 18, 2024
- **Time:** 14:50–15:20 (30 min)
- **Room:** Virtual Stage 1
- **Event:** World Congress 2024

## Description

In the rapidly evolving field of artificial intelligence, efficiently customizing large language models (LLMs) is essential for high performance across diverse applications. This talk will cover advanced techniques for parameter-efficient fine-tuning (PEFT) of LLMs, including p-tuning, prompt-tuning, adapters, supervised fine-tuning, and retrieval-augmented generation (RAG). Attendees will learn how to fine-tune LLMs with minimal computational resources while maintaining or enhancing performance, with practical demonstrations using the NVIDIA NeMo Framework. By the end, participants will be equipped to efficiently customize LLMs for specific tasks and challenges. This talk is valuable for researchers, developers, and AI enthusiasts interested in optimizing and personalizing LLMs.

## Speaker

### [Miguel Martínez](https://www.wearedevelopers.com/@miguel-martinez)

Senior Deep Learning Data Scientist at NVIDIA


## Related talks at this congress

- [Unlocking the Power of AI: Accessible Language Model Tuning for All](https://www.wearedevelopers.com/events/world-congress-2024/sessions/106-unlocking-the-power) — Cedric Clyburn, @technicallylegare
- [Efficient deployment and inference of GPU-accelerated LLMs​](https://www.wearedevelopers.com/events/world-congress-2024/sessions/73-efficient-deployment) — Adolf Hohl
- [Intro to LLMs and recent developments](https://www.wearedevelopers.com/events/world-congress-2024/sessions/397-intro-to-llms-and) — Christian Winkler
- [Leveraging Large Language Models for Legacy Code Translation: Challenges and Solutions](https://www.wearedevelopers.com/events/world-congress-2024/sessions/178-leveraging-large) — Michael Niebisch