LLM 튜토리얼

Learn how to use OpenAI AgentKit to automate GitHub issue reviews with AI agents that classify, detect duplicates, and summarize reports.

A practical guide to GPT-5 function calling: from JSON-schema functions to free-form tools, Lark/CFG constraints, preambles, and tool allowlists.

Learn about DeepSeek-V3.2-Exp, its new sparse attention mechanism, how it reduces API costs and improves long-context handling, and how to use it in your own projects.

Set up Ollama in Docker to run local LLMs like Llama and Mistral. Keep your data private, eliminate API costs, and build AI apps that work offline.

Learn how Anthropic's Imagine with Claude introduces a new paradigm for AI-assisted software development, generating functionality on the fly.

Learn how the Claude Agent SDK works by building three projects, from one-shot to custom-tool agents.

Learn how to run Qwen3-Next locally, serve it with transformers serve, and interact using cURL and the OpenAI SDK, making it ready for your app integrations.

Learn how Self-RAG improves traditional RAG by incorporating iterative reasoning and self-evaluation, and how to implement it step-by-step using LangGraph.

Compare LangChain, LangGraph, LangSmith, and LangFlow. Learn their roles, strengths, and when to use each for building production-ready AI applications.

Learn what LiteLLM is and how to use it for unified API calls to various LLM providers, covering basic API usage, error handling, fallbacks, streaming, structured outputs, and cost tracking.

Learn how to build a Streamlit app for side-by-side comparison of Qwen3-Next-80B-A3B and other Qwen models, focusing on real-world performance metrics.

Learn how to install and use Roo Code, an open-source AI coding assistant, and compare its features and functionalities with Cline.