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.
Genai Web is a web-based platform for government teams to manage generative AI applications, organize teams, and add apps built as external microservices.
Guide: Follow the documented GenAI Web setup sequenceGenquest Rag is a question-generation application that retrieves relevant information from Wikipedia and uses retrieval-augmented generation with a Weaviate vector store to create context-aware, answer-aware questions.
Guide: Generate a question from an answer-marked passageGo Qdrant Rag Sample is a Go application demonstrating semantic search and retrieval-augmented question answering over product catalog data, using OpenAI models and Qdrant.
Guide: Start the local catalog search APIGoki is a suite of Solana programs for key management and security, including a timelock-capable multisignature smart wallet and a token signer that authorizes transactions through NFT or token ownership.
Guide: Plan a staged transition of Goki signing controlGoogle ADK Training is a multi-module, hands-on course in a GitHub repository that teaches developers to build, test, deploy, and observe AI agents with Google Agent Development Kit (ADK).
Guide: Set up the course and begin Module 1Google Bard API is a Python FastAPI wrapper that exposes a REST endpoint for sending messages to Google Bard and receiving conversational responses.
Guide: Set up the local FastAPI serverGptqmodel is a Python toolkit for calibrating, quantizing, validating, saving, loading and running large language models.
Guide: Quantize and save a language modelHalo is an open-source Python framework for pre-training and post-training large language and multimodal models, including supervised fine-tuning, preference optimization, distillation and asynchronous multi-turn reinforcement learning.
Guide: Prepare a Halo training containerHugNLP is a Python library based on Hugging Face Transformers that provides models, data processors and application workflows for natural-language-processing research and development.
Guide: Install HugNLP from the repositoryHumanPrompt is a Python framework for researchers to design, manage, share and use human-readable prompts and prompt-based methods with large language models.
Guide: Set up HumanPrompt for a research experimentHummingbird is a C17 runtime for local inference of open-source dense and Mixture of Experts language models.
Guide: Build Hummingbird from sourceHybrid Search Rag is a Python package for building retrieval-augmented generation workflows, with documented hybrid retrieval methods, Milvus ingestion, MLflow tracking, Ragas evaluation and asyncio support.
Guide: Build and install the package wheelIgr is a Python deep-learning research implementation that learns implicit signed distance functions from raw 2D or 3D point clouds, with or without normals.
Guide: Train and mesh a single point-cloud surfaceInsurance Agentic AI is a Python insurance question-answering project with a Streamlit interface, a coordinator agent, and specialist agents for life, home and auto insurance.
Guide: Set up the project and launch its Streamlit interfaceJupiter V4 Arbitrage is an open-source Node.js and TypeScript command-line script intended to attempt token arbitrage on Solana through Jupiter Aggregator API v4, configured with a private key and RPC endpoint.
Guide: Prepare the script environmentKB-ALBERT is a Korean pretrained language model based on the lightweight ALBERT architecture, designed for financial-domain and general Korean NLP tasks. It provides a foundation that can be fine-tuned for downstream work.
Guide: Request KB-ALBERT model files for non-commercial useKosentencebert Skt is a Python project for training Sentence-BERT-style Korean sentence-embedding models using SKT KoBERT.
Guide: Train an embedding model with the documented scriptsKVQuant is a Python research project that reduces the memory footprint of long-context language-model inference by quantizing the model’s key-value cache to low precision.
Guide: Prepare KVQuant’s perplexity-reproduction workflow