xTuring is an open-source library for building, personalizing and controlling your own large language models, published on GitHub with about 2,700 stars. It makes fine-tuning open models such as LLaMA, Qwen and GPT-J simple, fast and cost-efficient, locally or in your private cloud.
xTuring covers the whole personalization pipeline with a simple API: dataset preparation, fine-tuning with techniques such as LoRA and mixed precision, evaluation, and inference with the adapted model. It supports a wide range of open model families, from GPT-2 and GPT-J to LLaMA, Mistral, Qwen3 and Mamba, and keeps data private by running entirely in your own environment rather than sending it to third-party APIs. The library is Apache-2.0 licensed, developed by Stochastic AI, and remains under active development in 2026. Documentation and examples cover preparing datasets in common formats, running distributed fine-tuning and exporting personalized models for deployment in downstream applications.
xTuring is completely free and open source under the Apache-2.0 license. There are no paid tiers for the library; you supply your own compute for training and inference.
xTuring is for developers and teams who want their own fine-tuned LLM adapted to their data and domain, without depending on proprietary fine-tuning APIs.