yxchng

LLaVA-pp

🔥🔥 LLaVA++: Extending LLaVA with Phi-3 and LLaMA-3 (LLaVA LLaMA-3, LLaVA Phi-3)

DoRA

[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation

ollama

Get up and running with Llama 3.1, Mistral, Gemma 2, and other large language models.

loraplus

LongMamba

Some preliminary explorations of Mamba's context scaling.

LinearAttentionArena

Here we will test various linear attention designs.

objaverse-xl

🪐 Objaverse-XL is a Universe of 10M+ 3D Objects. Contains API Scripts for Downloading and Processing!

DeepSeek-VL

DeepSeek-VL: Towards Real-World Vision-Language Understanding

Awesome_Mamba

Computation-Efficient Era: A Comprehensive Survey of State Space Models in Medical Image Analysis

flash-linear-attention

Efficient implementations of state-of-the-art linear attention models in Pytorch and Triton

DAQ-VS

Code For Our Work: DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries [ECCV-2024]

corenet

CoreNet: A library for training deep neural networks

evit

Python code for ICLR 2022 spotlight paper EViT: Expediting Vision Transformers via Token Reorganizations

CAT-Seg

Official Implementation of "CAT-Seg🐱: Cost Aggregation for Open-Vocabulary Semantic Segmentation"

inceptionnext

InceptionNeXt: When Inception Meets ConvNeXt (CVPR 2024)

EfficientVMamba

Code Implementation of EfficientVMamba

InternVL

[CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的可商用开源多模态对话模型

imagenet_d

[CVPR2024 Highlight] Official Code for "ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object"

LLM-Inheritune

This is the official repository for Inheritune.

croco

coconut_cvpr2024

VideoMamba

[ECCV2024] VideoMamba: State Space Model for Efficient Video Understanding

S-Seg

DeLVM

V3Det

open-eqa

OpenEQA Embodied Question Answering in the Era of Foundation Models

Diffusion-RWKV

Scaling RWKV-Like Architectures for Diffusion Models

RWKV-LM

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.

PEM

PEM: Prototype-based Efficient MaskFormer for Image Segmentation

frequency_determines_performance

Code for the paper: "No Zero-Shot Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance"