Adaptive Token Sampling for Efficient Vision Transformers (ECCV 2022 Oral Presentation)

A radically simple, reliable, and high performance template to enable you to quickly get set up building multi-agent applications

Code for "Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?" [ICML 2023]

Community Open Source Implementation of GPT4o in PyTorch

Implementation of the paper: "Audio Mamba: Bidirectional State Space Model for Audio Representation Learning" in pytorch

Deploy KoGPT with Triton Inference Server

HydraNet is a state-of-the-art transformer architecture that combines Multi-Query Attention (MQA), Mixture of Experts (MoE), and continuous learning capabilities.

RWKV Wiki website (archived, please visit official wiki)

Japanese NLP sample codes

reference pytorch code for huggingface transformers

Master's Final Degree Project on Artificial Intelligence and Big Data

Scripts run to produce the RiboTIE paper

O Gamei é uma aplicação open-source para geração de apresentações utilizando Inteligência Artificial, executando completamente local em seu dispositivo. Este projeto oferece controle total sobre seus dados e privacidade, permitindo o uso de modelos como OpenAI, Gemini e outros.

A collection of Jupyter Notebooks exploring key topics in Artificial Intelligence, including recommender systems, explainable AI, reinforcement learning, and transformers.

Deploy HuggingFace LLMs on AWS SageMaker with vLLM, OpenAI-compatible API (Lambda + API Gateway), and OpenWebUI chat interface. Default: Qwen2.5-1.5B-Instruct.

Production-grade, OpenAI-compatible server using vLLM v0.17.0. Deploy LLMs, Embeddings, or Vision models via environment variables.

A comprehensive collection of the 101 most useful commands for various programming languages, tools, and technologies.

Run local LLMs on Intel Arc Pro B60 GPUs using vLLM and Docker. Includes configuration guides and deployment tools for Linux environments.

Compute the score of similarity between two strings

Experimental ROCm 6.1 setup for early Qwen7B model tests on AMD GPUs. 🚧