$ show onlybooks/llm
This repository accompanies a Korean-language book on practical LLM application development, containing hands-on code examples organized by chapter. It demonstrates the full LLM lifecycle including foundational concepts (transformers), model optimization (quantization, PEFT), application building (RAG, agents, multimodal systems), and deployment techniques. The primary value lies in its educational notebooks that show how to implement LLM-powered solutions for tasks like Text2SQL, retrieval-augmented generation, and AI agents using frameworks such as LlamaIndex, AutoGen, and vLLM. While designed for Google Colab execution, the code illustrates reusable patterns for LLM engineering.
The LLM identified 3 files worth highlighting for reuse:
Provides a reusable asynchronous parallel API processor that handles rate limiting and retries for LLM API calls. Encapsulates complexity of managing concurrency and API constraints, allowing safe high-throughput LLM interactions with minimal external dependencies.
Contains utility functions for converting between JSONL and CSV formats, and generating structured evaluation requests for LLM output assessment. The conversion utilities are general-purpose, while the request generation follows a pattern adaptable to various evaluation frameworks, promoting reproducibility in LLM experimentation.
Demonstrates reusable patterns for loading quantized LLMs using Hugging Face Transformers with various methods (8-bit, 4-bit, GPTQ, AWQ). These concise configuration snippets significantly reduce memory footprint and are essential for deploying LLMs on resource-constrained environments, showcasing practical model optimization techniques.