Versatile audio super resolution (any -> 48kHz) with AudioSR.

Towards Human-Sounding Speech

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.

A Conversational Speech Generation Model

A simple and effective LLM pruning approach.

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

Training-free Post-training Efficient Sub-quadratic Complexity Attention. Implemented with OpenAI Triton.

Robust Speech Recognition via Large-Scale Weak Supervision

c-ray is a small, simple path tracer written in C

A basic Ray Tracer that exploits numpy arrays and functions to work fast.

A collaboration friendly studio for NeRFs

Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

SqueezeLLM: Dense-and-Sparse Quantization

PyTorch image models, scripts, pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNet-V3/V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeX…

FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/