Tools to Fine-Tune and Train AI Models Locally

Reef is self-hosted infrastructure for developers who want AI agents to improve through feedback on actual interactions. It connects inference and learning with versioned deployment, so an agent can update its model weights or its prompts, rules, and skills while continuing to serve requests. It's open source under Apache 2.0.

DiffSynth-Studio is a Python diffusion model engine for developers and researchers who want to generate media and train models on their own hardware. It supports large models on consumer GPUs through memory offloading and quantization, with inference and training in the same framework. It's open source under Apache 2.0.

Diffusers is an open-source Python library for developers and researchers who want to run diffusion models on their own hardware or build generation features into an application. It uses PyTorch and supports image, video and audio generation. The library is licensed under Apache 2.0 and supports Apple Silicon.

Gemma is Google DeepMind’s family of open-weight AI models for developers building applications that can run on their own hardware. Its range covers compact models for phones and IoT devices alongside larger Gemma 4 models for reasoning on personal computers and servers. Some applications can work offline, keeping model inference on the device. Google AI Studio and Google Cloud are also available for hosted use.

MLX is a machine learning array framework for researchers and developers building models on their own hardware. Its distinctive feature on Apple silicon is shared CPU and GPU memory: both processors can work on the same arrays without copying data between them. It's open source under the MIT license.

LLaMA-Factory is an open-source framework for developers and researchers who want to adapt language and multimodal models on their own hardware. It brings training and inference into one toolkit, with support for LLaMA, Qwen3, Qwen3-VL, DeepSeek, Gemma, Mistral and LLaVA. Its license is Apache 2.0.

Transformers is a Python library for developers and researchers who want to run pretrained AI models or train their own on hardware they control. It covers language, images, audio, video and multimodal work through a shared way of defining models. The library runs in a local Python environment; pretrained checkpoints are available from the separate Hugging Face Hub.

Unsloth brings model training and everyday AI use into a desktop app for people who want to run models on their own hardware. Its no-code interface covers chat, fine-tuning and media generation on macOS, Windows and Linux. The Unsloth software is open source under Apache 2.0.

Prodigy is a proprietary annotation tool that runs on your own machines, including air-gapped systems without an internet connection. It's for developers and research teams building training and evaluation datasets for custom AI models. The Python library includes a web application where annotators can label data without programming knowledge.

DataDreamer connects LLM prompting, synthetic data generation, and model training in one Python library. It's for researchers and developers who want to build datasets and use them to fine-tune or align models in reproducible workflows. The library is open source under the MIT license.

Seed-VC changes recorded speech or singing to sound like a voice supplied in a short reference clip, without training a separate model for that speaker. It runs locally on Windows, Linux and Apple Silicon Macs, with uses in audio production, live streaming and online meetings. The project is archived and no longer maintained.

so-vits-svc is an offline AI framework for changing the voice in an existing singing recording while preserving its pitch and intonation. It's aimed at developers and researchers who want to train their own singing voices, including fictional character voices. The project is archived and no longer maintained.

FluxGym is a local web interface for training FLUX LoRAs on your own images, with support for GPUs with 12GB, 16GB or 20GB of VRAM. It's for people who want to customize an image model through a browser while retaining access to detailed training controls. It runs on Windows and Linux, with Docker support, and is open source under the MIT license.

FlagEmbedding is an open-source Python toolkit for developers building semantic search or retrieval-augmented generation (RAG) into their own applications. It runs BGE embedding and reranking models, with tools to fine-tune both and evaluate retrieval results. The library uses the MIT license.

NMKD Stable Diffusion GUI is a local AI image generator for people who want to create and edit images on a Windows PC. It combines Stable Diffusion generation with inpainting, LoRA training and image post-processing in a desktop interface. It's open source under GPL-3.0.

DVC connects data and model versions to the code in your Git repository, so you can reproduce a machine learning experiment with the inputs it used. It's a free, open-source tool under Apache 2.0 for individual data scientists and small projects. It runs on macOS, Windows and Linux.

Aim is a free, open source ML experiment tracker for researchers and teams who want to keep training records on their own infrastructure. It runs in your training environment or on a self-hosted server, with Docker and Kubernetes deployment support. Its Apache 2.0 license permits use and modification.

DataTrove is an open-source Python library for teams preparing large text datasets, including LLM training corpora. It runs on your own machine or on Slurm and Ray clusters, with processing steps that carry across those environments. It uses the Apache 2.0 license.

nanoGPT is a Python toolkit for developers and researchers who want to train GPT models on their own hardware or fine-tune existing GPT-2 checkpoints. Its author has deprecated the project and points readers to nanochat. The MIT-licensed code remains available for study and modification.

TensorRT Model Optimizer, called NVIDIA Model Optimizer or ModelOpt, is a Python library for developers preparing models for local or self-hosted inference. It reduces model size and memory use and can speed up inference through compression and other optimization techniques. It's open source under Apache 2.0.

Ludwig is an open-source Python framework for developers and researchers who want to train custom AI models on their own hardware. A YAML file describes the model and training pipeline, while Ludwig handles preprocessing, training and evaluation. It uses the Apache 2.0 license. Install the Python package with the optional LLM dependencies for fine-tuning; current source requires Python 3.12 or later.

Intel Neural Compressor is a Python library for developers compressing AI models for deployment on their own hardware or servers. It supports local LLM work as well as other deep learning models, with particular attention to Intel CPUs, GPUs and Gaudi accelerators. It's open source under the Apache 2.0 license.

AutoAWQ is a Python library for developers who want to compress and run LLMs on their own hardware using 4-bit Activation-aware Weight Quantization (AWQ). The project is archived and no longer maintained. It's open source under the MIT license. It installs as a Python package, with optional kernel or Intel CPU dependencies.

H2O LLM Studio is a self-hosted tool for teams that want to adapt language models to their own datasets without writing training code. Its browser interface brings training experiments, evaluation, and model testing into one place. The project is open source under Apache 2.0.

StyleTTS 2 is an open-source text-to-speech model for developers and speech researchers who want to generate expressive speech on their own hardware. It can choose a speaking style from the text without a reference recording, while its multispeaker model uses reference audio to reproduce a speaker's voice and delivery. The Python code uses PyTorch and carries the MIT license.

AutoTrain trains custom machine learning models from your own data through a no-code interface. It's for people who need to fine-tune an LLM or build a classifier without writing a training pipeline. The local AutoTrain Advanced project is no longer maintained, so it won't receive bug fixes or new features.

Pyramid Flow is an open-source AI video generator for people who want to create clips on their own hardware, and for researchers working on video models. It turns text prompts into video or animates a supplied image with guidance from text. The Python code uses PyTorch and carries MIT. Model licenses are separate: the SD3-derived weights use the Stability AI Community License. Check the chosen checkpoint before deployment.

XTTS v2 generates speech from text using a reference voice recording or a preset speaker. It runs locally through Coqui TTS and suits developers building speech into apps, as well as researchers who want to fine-tune a speech model on their own hardware.

TensorZero is a self-hosted platform for developers building LLM applications. The project is archived and no longer maintained. It combines a model gateway with tools for inspecting responses, evaluating workflows, and improving prompts using production data and human feedback.

Open-Sora is an open source AI video generation project for developers, researchers, and creators who want to run and adapt a model on their own hardware. Its model focuses on turning reference images into video, with text prompts guiding the result. It also generates video directly from text. The code and Open-Sora 2.0 weights use Apache 2.0.

MNN is a lightweight C++ framework for developers who want AI models to run on phones, PCs and embedded devices. It handles inference and training on the device, with a focus on small application footprints and hardware acceleration. The project is open source under Apache 2.0, and Alibaba uses it in apps including Taobao, Youku and DingTalk.

DB-GPT is a self-hosted AI data assistant for teams analyzing business data and developers building data applications. It turns plain-language requests into SQL queries and Python analysis, then produces charts, dashboards, or HTML reports. You can run it on macOS or Linux, with Docker deployment also supported.

Kaito manages self-hosted LLM inference, fine-tuning, and document retrieval services in a Kubernetes cluster. It's for teams that want to run models on infrastructure they control while reducing the work of sizing GPU resources and managing model deployments. The project is open source under Apache 2.0.

Nemotron is NVIDIA's family of AI models for developers building agents that reason, write code and call tools. You can run models locally for private, offline work or deploy them on your own servers. NVIDIA publishes model weights, training data and recipes so teams can inspect and adapt the models for their applications.

Hallo turns a single portrait and a speech recording into an animated talking video on your own hardware. It's a local AI tool for creators working with talking portraits and researchers who want access to both generation and training code. The Python code uses the MIT license; required pretrained models and dependencies have their own terms.

Kolors is a text-to-image model for people who want to generate photorealistic images on their own hardware, including work with Chinese prompts and Chinese cultural content. Developed by Kuaishou, it understands prompts in Chinese and English and can render text in both languages within generated images.