bitsandbytes

By bitsandbytes Foundation ยท Fine-tuning and quantization

About bitsandbytes

A library for 8-bit and 4-bit model weights, used to load and fine-tune large models on smaller GPUs.

Details

  • Open source

    Yes

Not enough info yet on works offline after setup, local API server, runs on macOS, runs on Windows, runs on Linux. Check the tool's own site.

Listing written by LaptopMind, checked 2026-11-04. Report a correction or request removal. Back to fine-tuning and quantization.

More tools in fine-tuning and quantization

Unsloth

by Unsloth AI

A library for fine-tuning open language models faster and with less GPU memory, with ready notebooks.

  • Open source

Axolotl

by Axolotl AI

A configuration-driven tool for fine-tuning open models, from small adapters to full training runs.

  • Open source

MLX LM

by Apple

Tools built on MLX for running, quantizing and fine-tuning language models on Apple silicon Macs.

  • Open source
  • Runs on macOS

LLaMA-Factory

by hiyouga

A fine-tuning toolkit with a web interface that supports many open model families and training methods.

  • Open source