← back to the directory
Library / SDK Fine-tuning & Training

LitGPT

1Pretrain 2and 3finetune 4LLMs 5from 6scratch.

the six Ws · specification

W1 Who

Built by Lightning AI.

W2 What

Provides from-scratch implementations of many LLMs to pretrain and finetune.

W3 Where

A Python CLI and library on GPUs.

W4 When

When training or adapting an open model.

W5 Why

Readable, dependency-light recipes for twenty-plus models.

W6 With

PyTorch and Lightning Fabric.

W7 Watch

Apache-2.0, from Lightning AI; no credentials, runs on your hardware.

LightningPretrainingLoRA

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.