---
name: "llama-factory-unified-finetuning"
description: "Fine-tune, evaluate, and export 100+ LLMs with LLaMA-Factory unified interface."
metadata:
  author: "hiyouga"
  version: "1.0.0"
  runtime_compatibility:
    - "codex"
    - "claude-code"
    - "cursor"
    - "windsurf"
triggers:
  - "fine tune with llama-factory"
  - "run dpo alignment llama-factory"
  - "use llama-factory webui"
permissions:
  fs_read: true
  fs_write: "workspace"
  shell_exec: false
  net_access: false
---

# LLaMA-Factory Unified Fine-Tuning WebUI

Manages complete model alignment with LLaMA-Factory: pre-training, supervised fine-tuning (SFT), reward modeling, DPO/PPO alignment, WebUI visual trainer, and vLLM inference testing.

## When to Invoke
Invoke this skill whenever you need to:
- fine tune with llama-factory
- run dpo alignment llama-factory
- use llama-factory webui

## Execution SOP & Rules
1. Launch visual training dashboard with llamafactory-cli webui.
2. Configure dataset info inside data/dataset_info.json with prompt and response column mappings.
3. Execute training via command-line: llamafactory-cli train config.yaml.
4. Verify validation loss curves and export merged checkpoints ready for inference.

## Verification Checklist
- Run local linting, syntax validation, and type-checks before reporting completion.
- Ensure all created or modified files adhere to workspace boundaries and avoid uncommitted secrets.
- Verify tests or build commands exit cleanly with return code 0.