jaketae / koclip

Multimodal AI Story Teller, built with Stable Diffusion, GPT, and neural text-to-speech

N-gram keyword extraction using spaCy and pretrained language models

Ensembling Hugging Face transformers made easy

#自然语言处理#PyTorch implementation of FNet: Mixing Tokens with Fourier transforms

#计算机科学#[AAAI2024] FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning

Fast and memory-efficient exact attention

Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

Open-Sora： 完全开源的高效复现类Sora视频生成方案

Using Low-rank adaptation to quickly fine-tune diffusion models.

Official Pytorch implementation of "Visual Style Prompting with Swapping Self-Attention"

Official implementation of "ResAdapter: Domain Consistent Resolution Adapter for Diffusion Models".

现代 C++ 教程，中文书籍

Mora: More like Sora for Generalist Video Generation

#计算机科学#Official implementation for "Break-A-Scene: Extracting Multiple Concepts from a Single Image" [SIGGRAPH Asia 2023]

#大语言模型#Infinity is a high-throughput, low-latency serving engine for text-embeddings, reranking models, clip, clap and colpali

The simplest, fastest repository for training/finetuning medium-sized GPTs.

[CVPR 2024] FRESCO: Spatial-Temporal Correspondence for Zero-Shot Video Translation

Scene Text Recognition with Permuted Autoregressive Sequence Models (ECCV 2022)

[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.