xuanhan863

unsloth

Finetune Llama 3, Mistral & Gemma LLMs 2-5x faster with 80% less memory

gpt-fast

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

latent-consistency-model

Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference

Amphion

Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.

segment-anything-fast

A batched offline inference oriented version of segment-anything

tarsier

Vision utilities for web interaction agents 👀

stable-fast

Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.

PatchFusion

[CVPR 2024] An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation

punica

Serving multiple LoRA finetuned LLM as one

genmusic_demo_list

a list of demo websites for automatic music generation research

ziplora-pytorch

Implementation of "ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAs"

glake

GLake: optimizing GPU memory management and IO transmission.

SegDrawer

Simple static web-based mask drawer, supporting semantic segmentation with interactive Segment Anything Model (SAM) and video segmentation with XMem.

PoseAnything

Pengi

An Audio Language model for Audio Tasks

flash-fft-conv

FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores

Consistent4D

[ICLR 2024] Official Implementation of Consistent4D: Consistent 360° Dynamic Object Generation from Monocular Video

Diff-HierVC

Official Pytorch Implementation of "Diff-HierVC: Diffusion-based Hierarchical Voice Conversion with Robust Pitch Generation and Masked Prior for Zero-shot Speaker Adaptation"

diff-gaussian-rasterization-w-depth

flare

[SIGGRAPH Asia '23] FLARE: Fast Learning of Animatable and Relightable Mesh Avatars

d3po

[CVPR 2024] Code for the paper "Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model"

DepthFlow

🌊 Image to → 2.5D Parallax Effect Video. High quality, user first.

nxtp

Object Recognition as Next Token Prediction (CVPR 2024)

AQUA-Tk

AQUA-Tk = Audio QUality Assessment-Toolkit. (In development)

bandit

BandIt: Cinematic Audio Source Separation

MULTI-AUDIODEC

This is the official implementation of our multi-channel multi-speaker multi-spatial neural audio codec architecture.

DPI

ScorePerformer

ScorePerformer: Expressive Piano Performance Rendering with Fine-Grained Control (ISMIR 2023)

Meta-ControlNet

cordvox

experiments of fast, lightweight, high-quality neural vocoder for single speaker's speech synthesization