Coding & Development

Flash Diffusion + TAESD3 is a Hugging Face Space application designed for real-time Stable Diffusion 3 image generation. Users can input a text prompt, and the tool will generate a corresponding image. It leverages Flash Diffusion and TAESD3 technologies to achieve this. The application also allows for customizing the seed, which is useful for reproducibility of generated images. While the current live website indicates a runtime error, the intended functionality is to provide a platform for experimenting with and creating AI-generated images based on textual input.

Langchain-Chatchat, formerly Langchain-ChatGLM, is an open-source RAG (Retrieval Augmented Generation) and Agent application built upon the Langchain framework. It supports a wide array of large language models such as ChatGLM, Qwen, and Llama, and is designed for local, offline deployment. The tool facilitates knowledge-based question-answering by processing documents, segmenting text, vectorizing, and matching queries to generate responses using LLMs. It offers robust features including LLM and knowledge base dialogue, search engine integration, file RAG, and multi-modal image dialogue. Langchain-Chatchat also supports various model deployment frameworks like Xinference, Ollama, LocalAI, and FastChat, making it highly adaptable for different hardware and model types.

Lumina-mGPT 2.0 is an open-source, stand-alone, decoder-only autoregressive model designed for a broad spectrum of image generation tasks. Trained from scratch, it supports functionalities such as text-to-image generation, image pair generation, subject-driven generation, multi-turn image editing, controllable generation, and dense prediction. The project provides inference code for image-to-image tasks and all-in-one model checkpoints on HuggingFace. It also offers acceleration strategies like Speculative Jacobi Decoding and Model Quantization to optimize inference time and GPU memory usage. Lumina-mGPT 2.0 is ideal for AI researchers and machine learning engineers looking to explore and implement advanced image modeling techniques.

Luotuo-Chinese-LLM is an open-source initiative focused on advancing Chinese large language models. The project, developed by researchers from Huazhong Normal University and SenseTime, encompasses a range of models, datasets, pipelines, and applications. Key sub-projects include ChatHaruhi for character-based conversational AI, Luotuo Embedding for generative text embedding, Luotuo QA for conversational question answering, and Mini Luotuo for distilled instruction-following models. It also features Silk Road for building Chinese LLM data foundations and Silk Magic Book for collecting effective prompts. The project emphasizes practical applications and research into cross-language data tuning.

ICLR2025-Papers-with-Code is a comprehensive GitHub repository dedicated to compiling research papers and their corresponding open-source projects from the International Conference on Learning Representations (ICLR). The collection spans from ICLR 2021 to the upcoming ICLR 2025, with a particular emphasis on advancements in Large Language Models (LLMs) and various subfields within Natural Language Processing (NLP). This resource serves as a valuable hub for researchers, academics, and developers looking to stay updated on the latest research trends and access practical code implementations. The repository is actively maintained and updated, encouraging community contributions through watching, forking, and starring the project.

iir is an open-source project hosted on GitHub, offering a collection of algorithms and functionalities for machine learning, natural language processing, and information retrieval. Developed primarily in Python, Ruby, C++, and R, it serves as a valuable resource for researchers and developers in AI-related fields. The repository includes implementations for tasks such as active learning, clustering, natural language detection, LDA, PCA, perceptron, and various neural network components. Its modular structure allows users to explore and integrate different techniques for their specific AI projects, making it suitable for both academic research and practical application development.

GPT-Prompts is a GitHub repository dedicated to collecting and sharing useful prompts for various AI applications. The project is open-source, encouraging community contributions to expand its collection. A notable inclusion is the Midjourney Prompt Generator, which helps users create effective prompts for image generation. This resource is designed to assist individuals working with AI models by providing a centralized location for creative and functional prompts, fostering better interaction and output from AI tools. It serves as a practical resource for prompt engineering, allowing users to explore and utilize a range of prompts for different AI use cases.

LyCORIS is an open-source project designed for implementing diverse parameter-efficient fine-tuning algorithms specifically for Stable Diffusion models. Originating from LoCon, it offers a range of methods including LoRA, LoHa, LoKr, (IA)^3, DyLoRA, and Native fine-tuning (Dreambooth). This project provides flexible options for AI developers and machine learning engineers to customize and optimize their Stable Diffusion models. It supports integration with popular interfaces like a1111/sd-webui, ComfyUI, and InvokeAI, and offers multiple training methods including kohya-ss/sd-scripts and standalone wrappers for PyTorch modules. LyCORIS also includes utilities for extracting LoCon, merging models, and converting between different formats.

MLPP is a comprehensive C++ library designed to empower developers with robust machine learning capabilities. It bridges the gap between low-level C++ development and advanced machine learning engineering, offering a rich collection of algorithms for regression, deep neural networks, optimization, loss functions, regularization, and weight initialization. The library also includes modules for natural language processing, computer vision, numerical analysis, linear algebra, and statistics. With prebuilt neural networks, various datasets, and utility functions, MLPP provides a solid foundation for building and experimenting with machine learning models in C++.

MMagic (Multimodal Advanced, Generative, and Intelligent Creation) is an advanced and comprehensive open-source AIGC toolkit built on PyTorch, inheriting capabilities from MMEditing and MMGeneration. It offers state-of-the-art generative models for processing, editing, and synthesizing images and videos. Key applications include image restoration, text-to-image generation, 3D-aware generation, inpainting, matting, and super-resolution. MMagic supports fine-tuning for stable diffusion, ControlNet, GAN interpolation, and other popular GAN applications. Its efficient framework, based on MMEngine and MMCV, allows for modular construction of customized editor frameworks and provides rich components for controlling the training process.

Open Computer Use is an open-source AI agent designed to control a secure cloud Linux computer. Powered by E2B Desktop Sandbox and a variety of open-source LLMs, it allows AI to operate the computer using keyboard, mouse, and shell commands. The tool supports over 10 LLMs, including Llama 3.2/3.3, Groq, DeepSeek, Gemini 2.0 Flash, GPT-4o/4o mini, Claude, OS-Atlas, ShowUI, Moonshot, and Mistral AI, with an easy-to-integrate architecture for new models. It features live streaming of the sandbox display, allowing users to pause and prompt the agent at any time. While designed for Ubuntu, it can work with any operating system, providing a flexible and secure environment for AI-driven computer tasks.

Milnesium is a software company specializing in AI and Data Science solutions designed to tackle real-world business challenges. They provide machine learning-driven analytics to uncover data patterns, forecast trends, and power predictive models for demand, churn, and operational bottlenecks. Their Generative AI Suite offers AI co-pilots for tasks like drafting emails, summarizing text, and automating reports. Additionally, Milnesium provides Computer Vision and Multimodal AI techniques for interpreting visual content, intelligent document processing, and object recognition. The company emphasizes a blend of deep AI expertise with domain know-how, agile principles, and a proven track record in delivering custom AI and analytics solutions across various industries.

OS Ninja provides an intelligent way to explore and learn open-source projects by generating AI-powered learning paths for any repository. It decodes complex codebases and creates structured tutorials, diagrams, and documentation that evolve with the code. Users can search for open-source projects or request new ones to be added. The platform performs deep research on entire codebases, which can take up to 24 hours, to generate high-fidelity learning paths. It caters to various learning styles, including Socratic questioning, Feynman technique, and traditional book format. OS Ninja also offers curated collections of open-source repositories across categories like Generative AI, Data, Robotics, Game Engines, Crypto & Web3, and Machine Learning, making it a comprehensive resource for developers looking to master new codebases.

The National Centre of Artificial Intelligence (NCAI) at UET Lahore, established as the Al-Khwarizmi Institute of Computer Science (KICS) in August 2002, is dedicated to advancing research and development in Computer Science and Information Technology. KICS engages in various research activities, including the development of AI and machine learning solutions. The institute also emphasizes technology transfer through research labs, technology centers, and incubated startups. It actively participates in research collaborations and hosts conferences and workshops, such as the IEEE International Conference on Open Source System and Technologies (ICOSST), to disseminate knowledge and foster innovation in the AI domain.

Nota AI offers an AI model optimization platform called NetsPresso, designed to deploy high-performance AI on any device. The platform focuses on hardware-aware optimization, allowing for efficient on-device AI solutions. Nota AI provides various AI solutions including Nota Vision Agent, Industrial Safety Surveillance, Intelligent Transportation Systems (ITS), and Driver Monitoring Systems & Facial Recognition (DMS&FR). The company emphasizes unlocking the possibilities of AI through global collaboration and staying ahead with the latest AI insights, as demonstrated by their news and tech blog updates on topics like memory usage reduction for LLMs and commercialization expansion to data centers.

Agentic-Reasoning is an open-source tool designed to facilitate the integration of external tools into large language model (LLM) reasoning processes. This capability is crucial for developing advanced AI agents that can automate complex workflows by interacting with various systems and data sources. The tool provides a framework for developers and researchers to build sophisticated AI agents, enabling them to leverage diverse functionalities beyond the LLM's inherent knowledge. It is particularly useful for those working on collaborative AI projects, offering a structured approach to enhance agentic behavior and decision-making through tool utilization. By enabling LLMs to dynamically select and use tools, Agentic-Reasoning helps create more versatile and powerful AI solutions.

AIlice is a fully autonomous, general-purpose AI agent, designed to function as a standalone artificial intelligence assistant similar to JARVIS. Built on open-source LLMs, it utilizes a unique Interactive Agents Call Tree (IACT) architecture to break down complex tasks into dynamically constructed agents, integrating results with high fault tolerance. AIlice is proficient in tasks such as thematic research, coding, system management, and literature reviews, and aims for self-evolution where AI agents autonomously build feature expansions. It supports voice interaction, open-source and commercial models, native multi-modal capabilities, and rich media UI. Users can try AIlice online or install it locally, with options for GPU acceleration and specific feature installations like PDF reading or speech.

Opentensor Foundation is behind Bittensor, a groundbreaking decentralized network designed to foster an internet-scale machine learning ecosystem. This network operates on a token-based incentive model, rewarding miners for their computational contributions and knowledge sharing. By distributing value directly to contributors, Bittensor aims to create a pure and open market for artificial intelligence, free from central control. This approach encourages broad participation and innovation, allowing for the collective development and deployment of AI models and services. The platform's core mission is to democratize AI, making advanced machine learning accessible and beneficial to all participants.

BELLE, which stands for "Be Everyone's Large Language model Engine," is an open-source initiative by LianjiaTech focused on advancing Chinese dialogue large language models. Unlike projects primarily concerned with pre-training, BELLE emphasizes enabling individuals to create their own high-performing, instruction-following language models based on existing open-source pre-trained models. The project continuously releases instruction training data, relevant models, training code, and application scenarios. It also evaluates the impact of different training data and algorithms on model performance, with a specific optimization for Chinese language using ChatGPT-generated data. Recent updates include enhanced Chinese speech recognition models, multimodal large language models, and research reports on fine-tuning strategies and RLHF training.

Aikido Security is a comprehensive platform designed to secure code, cloud, and runtime environments. It offers a unified approach to security, integrating features like Software Composition Analysis (SCA), Static Application Security Testing (SAST), Infrastructure as Code (IaC) scanning, and Dynamic Application Security Testing (DAST). The platform aims to reduce noise by contextualizing vulnerabilities and filtering out false positives, claiming a 95% reduction in alerts. Key capabilities include AI-powered continuous pentesting, automated vulnerability fixing with generated pull requests, and real-time runtime protection. Aikido also provides features for compliance, vulnerability management, and supply chain safety, making it a robust solution for developers and security teams.

Awesome-ChatGPT-prompts-ZH_CN is a GitHub repository dedicated to providing a diverse collection of Chinese prompts for ChatGPT and other large language models like Claude. It offers creative methods to customize AI behavior, such as transforming ChatGPT into a 'cat-girl' persona for role-playing scenarios. The repository also includes techniques for bypassing certain AI content restrictions and limitations, with specific instructions for ChatGPT and NewBing. It features tools for exporting conversations, bypassing WAF errors, and enhancing the AI's mathematical capabilities. The project is open-source, encouraging community contributions and providing updates on new bypass methods and prompt engineering techniques.

awesome-chatgpt-zh is a comprehensive, open-source Chinese guide designed to empower users with the knowledge and resources to effectively leverage ChatGPT. Hosted on GitHub, this project offers detailed instructions, prompt engineering guidelines, and application development insights. It curates a wide array of free and paid ChatGPT resources, along with a list of top open-source projects and productivity tools built on ChatGPT's capabilities. The guide covers various aspects, from understanding what ChatGPT is to advanced topics like LLM development, RAG guidance, and AGI concepts, aiming to significantly boost user productivity.

zzz-api offers a robust and stable API interface for accessing a wide range of large language models, including OpenAI, Anthropic Claude, Google Gemini, xAI Grok, and Chinese models like Baidu Wenxin Yiyan and Alibaba. It functions as an OpenAI API proxy, supporting advanced features such as video, batch processing, assistants, fine-tuning, and models like GPT-4o, GPT-5, and Sora-2. A key advantage is the elimination of the need for an OpenAI key, account, or US bank card, simplifying access for developers and enterprises. The service is compatible with OpenAI's API format, allowing for seamless integration and replacement of official OpenAI endpoints. It also supports streaming, embeddings for Langchain and vector databases, DALL-E-3 for image generation, Whisper for speech recognition, and TTS for voice synthesis.

deep_learning is an open-source GitHub repository that serves as a collection of summarized notes and experimental results in deep learning. It is designed for non-professional NLP and image engineers, as well as those new to deep learning, to quickly get started and implement solutions. The repository covers a wide range of topics including Bert, CNN, DeepFM, DeepInterestNetwork, Doc2Vector, ESSM, MLPs, RCNN_GRU, RNN, SSD, TextCNN, Wide & Deep, XDeepFm, and YoutubeNetwork. It aims to provide practical insights and code examples for various deep learning applications, making it easier to understand and apply complex models in real-world scenarios.