# stochasticai/xTuring

Build, personalize and control your own LLMs. From data pre-processing to fine-tuning, xTuring provides an easy way to personalize open-source LLMs. Join our discord community: https://discord.gg/TgHXuSJEk6

Repository: https://github.com/stochasticai/xTuring
Canonical: https://ross.abutalabs.com/products/xturing
Homepage: https://xturing.stochastic.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-learning, fine-tuning, gpt-2, gpt-j, llama, llm, lora, language-model, finetuning, adapter, gen-ai, generative-ai, mistral, mixed-precision, peft, quantization
Last push: 2026-03-04T23:07:06+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 70, release rhythm 8, longevity 90
- inputs: {"age_days": 1263, "days_push": 182, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2674, forks 211 (observed 2026-08-28T04:07:09.664348+00:00)

## What it is
xTuring is a Python library for fine-tuning, evaluating, and running open-source large language models such as LLaMA, GPT-J, GPT-2, Qwen, and GPT-OSS on your own data. It offers a simple API covering dataset preparation, memory-efficient fine-tuning (LoRA, INT8/INT4), inference, and model saving/loading.

## Use cases
- fine-tune llama on my own data
- personalize an open-source LLM with LoRA
- instruction-tune a model with an alpaca dataset
- run memory-efficient LLM fine-tuning on a single GPU
- quantize a language model to int4 or int8
- evaluate fine-tuned LLMs with perplexity
- fine-tune gpt-oss or qwen3 locally

## When to choose
- you want a simple, few-lines-of-code API to fine-tune open-source LLMs
- you need LoRA or low-precision (INT8/INT4) fine-tuning to cut hardware costs
- you want to keep training and inference private on your own machine or VPC
- you need to scale from laptop CPU to multi-GPU training

## When to avoid
- you need full pretraining of large language models from scratch
- you require a managed fine-tuning service with hosted infrastructure
- you need fine-tuning support for non-text modalities like images or audio
- you depend on cutting-edge model support the day a model is released

## Facets
- artifact type: library
- maturity: active
- function: llm-training, machine-learning, deep-learning, llm-inference, sdk
- domain: large-language-models, machine-learning, deep-learning, artificial-intelligence
- platform: python, cross-platform
- tags: fine-tuning, lora, peft, quantization, instruction-tuning, generative-ai, gpu

## Member repositories
- stochasticai/xTuring (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:09.664348+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:18:04.465383+00:00, confidence not recorded.
  - readme: https://github.com/stochasticai/xTuring (fetched 2026-08-28T04:07:09.664348+00:00, sha aaf302d17947)
  - homepage: https://xturing.stochastic.ai (fetched 2026-08-29T10:00:44.904717+00:00, sha 6617fef4f110)
  - site_page: https://xturing.stochastic.ai/overview/installation (fetched 2026-08-29T10:00:44.913780+00:00, sha e425ea0b0ac4)
  - site_page: https://xturing.stochastic.ai/overview/quickstart (fetched 2026-08-29T10:00:44.915455+00:00, sha 76fba045f339)
  - site_page: https://xturing.stochastic.ai/overview/quickstart/load_save_models (fetched 2026-08-29T10:00:44.917448+00:00, sha 4c66b0a9f6a1)
  - site_page: https://xturing.stochastic.ai/overview/quickstart/prepare (fetched 2026-08-29T10:00:44.919356+00:00, sha 9777019995a5)
  - site_page: https://xturing.stochastic.ai/overview/quickstart/data_usage (fetched 2026-08-29T10:00:44.921166+00:00, sha 1fb820908dda)
  - site_page: https://xturing.stochastic.ai/overview/quickstart/finetune_guide (fetched 2026-08-29T10:00:44.923028+00:00, sha 7dda161dd648)
  - site_page: https://xturing.stochastic.ai/overview/quickstart/inference (fetched 2026-08-29T10:00:44.924771+00:00, sha e7fc4b6f00ea)
  - registry_pypi: https://pypi.org/pypi/xturing/json (fetched 2026-08-29T10:00:44.938882+00:00, sha c604ff0b2155)
- Data as of 2026-08-30T08:39:29.467469+00:00.