AMD Supported Open-Source Models

PyTorch is a popular machine learning and deep learning framework that is based on Python and the Torch library.

TensorFlow is a popular machine learning and deep learning framework that was initially developed by the Google Brain team.

JAX is a Python library for accelerator-oriented array computation and program transformation, designed for high-performance numerical computing and large-scale machine learning.

vLLM is a library for high throughput, memory efficient large language model inference and serving.

SGLang is a comprehensive framework developed for efficient serving of large-scale language and vision-language models.

Hugging Face PEFT is a library for efficiently adapting large pretrained models to downstream applications by fine tuning a small number of model parameters.

Hugging Face TGI is a toolkit for deploying and serving Large Language Models (LLMs).

Hugging Face Transformers is a library providing easy access to thousands of pretrained state-of-the-art models.

OpenAI Triton is a programming language and compiler for writing highly efficient custom GPU kernels. 

OPEA is an open platform project for the creation of open, multi-provider, robust, and composable GenAI solutions for enterprise AI.

The bitsandbytes library is a lightweight Python wrapper for 8-bit and 4-bit quantization functions.

LlamaIndex is a framework for building LLM applications that can efficiently interact with specific data.

Ollama is a platform for building and running large language models on your local machine.

ComfyUI is a flexible, easy-to-use user interface (UI) framework for the design and execution of efficient stable diffusion pipelines.