Skip to content

AMD-Llama-135m

A LLaMA2 based model for text and code

Visit official site

What AMD-Llama-135m is

The model is described as a language model trained on the LLaMA2 architecture that runs on an AMD MI250 GPU, generating text and code for natural language processing work.

What you can do with it

  • Generate text from prompts
  • Generate code for programming tasks
  • Load the model on Hugging Face
  • Pretrain or fine tune the model

Application scenarios

  • Generate story text
  • Assist with writing code
  • Support natural language processing research

Getting started

  • Load the model from Hugging Face
  • Encode text with AutoTokenizer
  • Generate and then decode text or code

Pros & cons

✓ What we like

  • Pretraining and fine tuning are supported
  • It uses the same tokenizer as LLaMA2

! What to watch out for

  • The architecture follows LLaMA2
  • The described setup uses an AMD MI250 GPU

FAQ

What can it generate?

Text and code are listed as outputs.

Where is it hosted?

Loading and use through Hugging Face are described.

Last reviewed: 2026-11-15

More LLM API platform tools

View all →

How we review