axolotl

Configure and run YAML-driven LLM fine-tuning with Axolotl.

2|Updated Apr 25, 2026

Search Tags:#lora #dpo #rlhf #grpo #llm-fine-tuning #yaml-config #axolotl

Ready to run

npx skills add https://github.com/john-data-chen/hermes-agent-backup --skill axolotl-john-data-chen

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: axolotl
Source: https://github.com/john-data-chen/hermes-agent-backup/tree/main/skills/mlops/training/axolotl
Command: npx skills add https://github.com/john-data-chen/hermes-agent-backup --skill axolotl-john-data-chen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires axolotl, torch, transformers, datasets, peft, accelerate, deepspeed, and includes references (resource) components.

What problem does it solve?

Axolotl enables YAML-configured fine-tuning of large language models, making model customization reproducible and accessible without deep coding.

Core Features & Use Cases

Fine-tuning LLMs with LoRA/QLoRA, DPO, and GRPO pipelines.
Templates for training configurations across SFT and RLHF workflows.
Use Case: quickly prototype a tuning job by writing a YAML spec and launching it with the agent.

Quick Start

Create and run a YAML-based Axolotl fine-tuning configuration to train an LLM with LoRA, DPO, or GRPO.

Frequently Asked Questions about axolotl

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I configure LLM fine-tuning with YAML for LoRA and DPO workflows?▼

Axolotl enables YAML-configured fine-tuning of large language models, making model customization reproducible and accessible without deep coding. It supports LoRA/QLoRA, DPO, and GRPO pipelines through simple YAML files.

What is the best way to start an SFT or RLHF training job without deep coding?▼

The best way to start SFT or RLHF training is using YAML-driven configurations to quickly prototype a tuning job. You can define training parameters in a YAML spec and launch it directly with the agent.

Does axolotl work with DeepSpeed and PEFT for large language model training?▼

Yes, axolotl works with DeepSpeed and PEFT for large language model training. It relies on dependencies including torch, transformers, datasets, accelerate, and deepspeed to execute its YAML-driven fine-tuning pipelines.

Can I use GRPO and RLHF pipelines by only modifying a YAML configuration file?▼

Yes, you can run GRPO and RLHF pipelines by modifying a YAML configuration file. The skill provides templates for training configurations across both SFT and RLHF workflows via simple YAML specs.

Do I need PyTorch and Transformers installed to run YAML-driven LLM tuning jobs?▼

Yes, you need PyTorch and Transformers installed to run YAML-driven LLM tuning jobs. Required dependencies for executing the fine-tuning pipelines include torch, transformers, datasets, peft, accelerate, and deepspeed.