主要记录大语言大模型（LLMs） 算法（应用）工程师相关的知识及面试题

本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）

leaked prompts of GPTs

Generative Models by Stability AI

GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.

A super-fast Python implementation of seam carving algorithm for intelligent image resizing.

Code for CVPR 2019 paper. BASNet: Boundary-Aware Salient Object Detection

Official repository for CF-Font: Content Fusion for Few-shot Font Generation.

A series of large language models developed by Baichuan Intelligent Technology

Firefly: 大模型训练工具，支持训练Qwen2.5、Qwen2、Yi1.5、Phi-3、Llama3、Gemma、MiniCPM、Yi、Deepseek、Orion、Xverse、Mixtral-8x7B、Zephyr、Mistral、Baichuan2、Llma2、Llama、Qwen、Baichuan、ChatGLM2、InternLM、Ziya2、Vicuna、Bloom等大模型

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Best practice for training LLaMA models in Megatron-LM

Official repository for "PosterLayout: A New Benchmark and Approach for Content-aware Visual-Textual Presentation Layout" (CVPR 2023).

Official implementation of the ICASSP-2022 paper "Text2Poster: Laying Out Stylized Texts on Retrieved Images"

RealSR super resolution implemented with ncnn library

Real-World Super-Resolution via Kernel Estimation and Noise Injection

MAT: Mask-Aware Transformer for Large Hole Image Inpainting

Wide-Context Semantic Image Extrapolation, CVPR2019

Make HD zoom video of stable diffusion outpainting

Kandinsky 2 — multilingual text2image latent diffusion model

以JianYangAI作后端，进行在线雀魂对局

自动化雀魂AI的SDK，实时解析雀魂对局信息，并模拟鼠标动作出牌

Mjx: A framework for Mahjong AI research

日本麻将助手：牌效+防守+记牌（支持雀魂、天凤）

thumbor is an open-source photo thumbnail service by globo.com

stable diffusion webui colab

Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) with Stable Diffusion

fast-stable-diffusion + DreamBooth

Fengshenbang-LM(封神榜大模型)是IDEA研究院认知计算与自然语言研究中心主导的大模型开源体系，成为中文AIGC和认知智能的基础设施。

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.