QLoRA
TrainingQuantized Low-Rank Adaptation. A LoRA variant that quantizes the base model to 4 bits (NF4) before training: it lets you adapt models up to 65B on a single 48 GB GPU, with adapters of a few MB.
Quantized Low-Rank Adaptation. A LoRA variant that quantizes the base model to 4 bits (NF4) before training: it lets you adapt models up to 65B on a single 48 GB GPU, with adapters of a few MB.