LLM Engineer Toolkit
A curated, categorized list of over 120 LLM libraries, covering training, application development, RAG, inference, serving, data extraction and generation, agents, evaluation, monitoring, prompts and safety.
LLM Engineer Toolkit is a curated list of more than 120 libraries for working with large language models, organized by category. It links out to libraries for LLM training and fine-tuning, application development, retrieval-augmented generation, inference, serving, data extraction, data generation, agents, evaluation, monitoring, prompts, structured outputs, safety and security, and embedding models.
The training and fine-tuning section, for example, lists libraries such as Unsloth, PEFT, TRL, Transformers, Axolotl and LitGPT, each with a short description of what it does. The repository also links to related resources: a collection of LLM interview questions, a set of prompt-engineering techniques with LangChain implementations, and a collection of survey papers on LLMs, RAG and agents.