# Hugging Face — Make LLM Fine-tuning 2x faster with Unsloth and 🤗 TRL

- Company: Hugging Face (huggingface.co)
- Announced: 2024-01-10T00:00:00+00:00
- Category: developer-tool-release
- Subject: Platform
- Models affected: Llama, CodeLlama, Yi, Mistral, Code Llama 34b, Llama-2 7b, Mistral 7b, Tiny Llama 1.1b, Zephyr, Deepseek, TinyLlama, Llamafied Qwen, Llama-2 13b, codellama-34b
- Source: https://huggingface.co/blog/unsloth-trl
- Record: https://forck.live/items/1961-make-llm-fine-tuning-2x-faster-with-unsloth-and-trl

Hugging Face announces the integration of Unsloth with TRL, enabling faster LLM fine-tuning with up to 2x speedup and reduced memory usage, supporting Llama and Mistral architectures.

## Evidence

Verbatim from https://huggingface.co/blog/unsloth-trl:

> Make LLM Fine-tuning 2x faster with Unsloth and 🤗 TRL

---

Record: https://forck.live/items/1961-make-llm-fine-tuning-2x-faster-with-unsloth-and-trl
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md