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

XTuner

1Efficiently 2finetune 3large 4models 5and 6MoEs.

the six Ws · specification

W1 Who

Built by InternLM (Shanghai AI Lab).

W2 What

A training engine for QLoRA and full fine-tuning of large and MoE models.

W3 Where

A Python toolkit on GPUs.

W4 When

When fine-tuning on limited GPU memory.

W5 Why

Fine-tunes big models on small hardware.

W6 With

PyTorch and DeepSpeed.

W7 Watch

Apache-2.0, from InternLM; no credentials, runs on your GPUs.

InternLMQLoRAMoE

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.