Fine-Tuning Studio

PEFT / LoRA Adapter Training Engine · CPU Accelerated

Datasets (0)

Loading datasets...

QLoRA / PEFT

Adapter Output Name

Base Model (Local PyTorch)

Scanning models...

Training Preset

CPU Execution Profile: Training uses float32/QLoRA weights with gradient accumulation. Expect 1-3 min per epoch for small datasets.

Idle

No active training session.

Download a PyTorch model from Model Manager to begin.