Triton implementation of FlashAttention2 that adds Custom Masks.

Conan - The open-source C and C++ package manager

JSON for Modern C++

Development repository for the Triton language and compiler

Hackable and optimized Transformers building blocks, supporting a composable construction.

Clean, minimal, accessible reproduction of DeepSeek R1-Zero

LLM inference in C/C++

Fully open reproduction of DeepSeek-R1

Run any open-source LLMs, such as DeepSeek and Llama, as OpenAI compatible API endpoint in the cloud.

LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

A high-throughput and memory-efficient inference and serving engine for LLMs

Evaluate your speech-to-text system with similarity measures such as word error rate (WER)

Verbatim Automatic Speech Recognition with improved word-level timestamps and filler detection

A parallel implementation of gzip for modern multi-processor, multi-core machines.

Class to simplify communication with WeedFS

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

PyTorch-MPI-DDP-example

A Vectorized Python Dict/Set

Large, modern dataset for speech recognition

This package aims at simplifying the download of the AudioSet dataset.

Inference code for PaSST, using the HEAR API.

Efficient Training of Audio Transformers with Patchout

Scenic: A Jax Library for Computer Vision Research and Beyond

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)

Code for the Interspeech 2021 paper "AST: Audio Spectrogram Transformer".

Code for the Interspeech 2021 paper "AST: Audio Spectrogram Transformer".

A free audio dataset of spoken digits. An audio version of MNIST.

Torch implementation of ViT based classifier for Audio classification