Explore AI, agents, blockchain and crypto through official sources, concrete tasks and practical guides.
Selected profiles with a clear purpose, supported use cases and source-based guides. Selection is not an execution test or a ranking of project quality.
An open-source, self-hosted AI engine and API layer for running models locally across language, vision, voice, image, and video tasks.
Builder toolsOpenAI Agents SDKAn open-source, MIT-licensed Python framework for building and coordinating multi-agent LLM workflows. It supports configurable agents, tool use, handoffs and tracing.
InfrastructureLight ProtocolAn open-source Rust protocol for ZK compression on Solana. Its stated purpose is to reduce the cost or footprint of Solana state using zero-knowledge compression.
Builder toolsVelora SDKAn open-source TypeScript SDK for calling the Velora API and accessing aggregated decentralized-exchange liquidity and rates.
Selected for a documented reader task. Source review does not claim an execution test.
Drqa is a web-based PDF question-answering application codebase that converts documents into text chunks, embeds and retrieves relevant passages with Qdrant, and uses an LLM such as OpenAI GPT-3 to generate answers.
Guide: Set up the Drqa applicationDrt is a public research repository that brings together model checkpoints, data and papers for machine translation systems using deep reasoning.
Guide: Generate a translation with DRT-7B and TransformersElegantrl is a Python deep reinforcement-learning framework for implementing model-free algorithms and training agents in simulated environments, including workflows built around parallel simulation.
Guide: Install ElegantRL prerequisites for environment-based trainingEmotube is a Dockerized web application for analyzing sentiment in YouTube comments, with a React and TypeScript interface, a FastAPI backend, Transformers and PyTorch for sentiment analysis, and PostgreSQL storage.
Guide: Set up Emotube with Docker ComposeEnd To End Graphrag Implementation is a GitHub repository guide for configuring GraphRAG with Azure OpenAI, turning text files into entity-and-relationship graph artifacts, and querying the indexed material with local or global methods.
Guide: Set up and query a GraphRAG projectEnterprise Rag Framework is a Python framework for building retrieval-augmented generation applications that search organizational documents and provide relevant context to language models.
Guide: Index documents and query them with the Python APIEntity Db is a JavaScript library that stores vectors in browser IndexedDB and supports similarity search, with Transformers.js for generating text embeddings.
Guide: Install Entity Db and initialize a browser databaseEntroPE is an open-source Python reference implementation for long-term time-series forecasting.
Guide: Set up the EntroPE forecasting environmentEpipolarpose is an open-source PyTorch implementation of a self-supervised method for 3D human pose estimation.
Guide: Prepare the repository for its demo notebookEsmdiff is a Python research codebase for Structure Language Models, an SE(3)-invariant framework for protein conformation generation.
Guide: Install Esmdiff and configure ESM3-open
Ethereum is a programmable blockchain. Its documentation connects accounts, transactions, smart contracts, nodes and client APIs in one development stack.
Guide: Choose an Ethereum development pathF The Docs is a Python command-line application that parses documentation from text or JSON files, embeds it into a collection, and retrieves relevant passages in response to questions.
Guide: Load the included text example and open asking modeFashn Human Parser is a Python computer-vision model based on SegFormer-B4 that segments people in images into 18 semantic classes, including body parts, garments and accessories.
Guide: Install the Python package and prepare for inferenceFASHN VTON v1.5 is a Python virtual try-on model and inference pipeline that generates an image of a person wearing a supplied garment, without requiring segmentation masks.
Guide: Run a local try-on inferenceFbow is a C++ computer-vision library that builds visual vocabularies from image collections and transforms binary or real-valued image descriptors into bag-of-words representations.
Guide: Use the included ORB vocabulary for descriptor processingFiftyone Multimodal Rag Plugin is a Python plugin for FiftyOne that provides operators to create datasets from multimodal documents, index them with LlamaIndex and Milvus, and query retrieved image and text context for RAG responses.
Guide: Create and query a multimodal RAG indexGEMMA3 Object Detection is a Python repository demonstrating how to fine-tune Gemma 3 for object detection by representing bounding boxes as discrete location-token sequences.
Guide: Set up the fine-tuning environmentGemma3Moe Builder is a Python toolkit that sparse-upcycles dense Gemma-3-270m checkpoints into routed mixture-of-experts models, supports training and routing inspection, and provides a llama.cpp CPU-serving path using GGUF.
Guide: Convert a Gemma-3-270m checkpoint into an MoE