LLM Fine-tuning and Quantization

6 repos

Libraries and frameworks for efficiently fine-tuning large language models through quantization, parameter-efficient methods, and memory optimization techniques. These tools enable practitioners to adapt pretrained models to specific tasks and domains while reducing computational requirements and memory footprint. The cluster centers on technologies like QLoRA and bitsandbytes that make training accessible on consumer hardware, alongside training orchestration frameworks like Axolotl that integrate these techniques into end-to-end workflows.

Related clusters