Llama Factory

If you're fine-tuning LLMs and want to avoid writing boilerplate training code, this gives you expert guidance on LLaMA-Factory's WebUI and APIs. It covers the no-code interface, QLoRA configurations from 2-bit to 8-bit, and multimodal support across 100+ models including Llama, Qwen, and Gemma. The references include getting started guides, advanced techniques, and API docs pulled from official sources. Honestly most useful if you're already committed to LLaMA-Factory as your training framework. The skill helps you navigate features and debug issues without constantly context-switching to documentation.

Install to Claude Code

npx -y skills add orchestra-research/ai-research-skills --skill llama-factory --agent claude-code