Streamlined fine-tuning framework for LLMs. Supports full fine-tune, LoRA, QLoRA, FSDP, DeepSpeed, and multi-GPU. YAML config driven. Works with Llama, Mistral, Qwen, DeepSeek, and hundreds of HF models.
SkillScore86Maintained
Automatic
◆AI Engineer pack
Last verified 8/21/2026
What it changes
Passed 19 of 22 test cases with the skill, and 12 of 22 without it.
Without skill55%
With skill86%
Checks it passed that the model alone missed
The config sets lora_r to 16
The config sets lora_alpha to 32
The config specifies num_epochs as 3
and 6 more in the full report
7 fixed12 passed either way3 still failing0 broke
Pass-rate lift
+32 pp
Tokens
-21%
Agent turns
0%
Response time
-53%
Model gemini-3.6-flashMeasured 8/21/2026v119 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 7
Prompt
What rank dimension integer value is set for lora_r in the default Axolotl configuration template?
What was checked, without and with the skill
Checked
Without skill
With skill
The config sets lora_r to 16
Missed
Passed
case-11 · gemini-3.6-flash · 8/21/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md54 linessingle file
Overview
Axolotl is a fine-tuning framework supporting SFT, QLoRA, LoRA, full fine-tuning, DPO, and multimodal tuning for 100+ models (Llama, Mistral, Qwen, Gemma, DeepSeek). YAML-driven config avoids boilerplate. Supports multi-GPU, FSDP, DeepSpeed, and flash attention.
Installation
git clone https://github.com/OpenAccess-AI-Collective/axolotl
cd axolotl
uv pip install -e .
BentoML — model serving and deployment. Build prediction services from any ML framework with OpenAPI/Swagger. Containerize, deploy to Kubernetes, AWS, GCP, Azure. Adaptive batching and GPU support.