Tools to Fine-Tune and Train AI Models Locally

Trainable Weka Segmentation is a local image segmentation plugin and Java library for Fiji. It's for researchers and image analysts who want to teach a classifier which image regions belong to each class, then apply that model to other images or stacks. It's free and open source under GPL-3.0.

MFLUX is a Python toolkit for creators and developers who want to generate and edit images on an Apple Silicon Mac. It provides a command-line interface for working with models locally.

Supervision is a Python library for developers building applications around computer vision models. It supplies tools for working with predictions and datasets, so you don't have to build those parts of an application yourself.

PixlStash is an image manager for photographers culling shoots, creators reviewing AI images and people preparing datasets. It helps you find relevant pictures and review which ones to keep in large collections.

SimpleTuner is a diffusion model training toolkit for creators and researchers adapting image, video or audio models to their own datasets. Its web dashboard also supports teams sharing GPU workers.

BoTorch is a Python library for researchers and experienced AI practitioners who build Bayesian optimization algorithms. It helps choose which inputs to evaluate next by fitting a probabilistic model to observed results. Its main appeal is control over the models and decision rules used in that search.

bitsandbytes is a Python library for developers who need to run or fine-tune large language models on their own hardware with less memory. It works within PyTorch workflows.

Kolors generates photorealistic images from text for creators and developers who want to run image generation on their own hardware. Developed by Kuaishou, it supports Chinese and English prompts, including descriptions of Chinese cultural content.

RKLLM is a toolkit for developers building local AI applications on Rockchip boards. It runs language and multimodal models on the chip's neural processing unit (NPU).

Keras is an open-source Python deep learning framework for developers and researchers who want to build and train models without tying their code to one backend. It runs locally on Linux and macOS, with WSL2 recommended for Windows. Its Apache 2.0 license permits commercial use and modification.

DB-GPT is a self-hosted AI data assistant for teams analyzing business data and developers building data applications. It connects plain-language questions to database queries and code-based analysis.

Model2Vec is a Python package for developers who need text embeddings for search or classification on their own hardware. It turns sentence transformers into smaller static embedding models, trading a small drop in performance for lower model size and faster CPU inference.

ExLlamaV2 and V3 are local LLM inference libraries for developers building AI applications or hosting models on consumer GPUs. V2 is archived for now and isn't currently maintained; development continues in V3.

Data-Juicer is a Python framework for researchers and teams preparing datasets for AI training, evaluation, agents or document retrieval. It processes data on your own machine or across a Ray cluster.

XTTS v2 generates speech from text on your own hardware, using a voice recording as a reference. It's for developers adding speech to applications and researchers adapting a speech model to their own data.

Stanza is a Python NLP library for developers and researchers who need to analyze text in multiple human languages on their own hardware. It turns raw text into linguistic annotations, such as word forms, grammatical relationships, and named entities. It's open source under Apache 2.0.

StyleTTS 2 is a 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.

Unsloth is a desktop app for people who want to chat with models, train them on their own data, and create media on their own hardware. Its no-code interface brings these tasks into one workspace.

AI Toolkit (ostris) is a training suite for people fine-tuning image and video generation models on their own hardware or rented servers. It provides a browser interface and command-line access to training jobs.

diffusion-pipe trains image and video diffusion models on your own GPU hardware. It's for people fine-tuning models that are too large to fit on a single GPU or who want to distribute training across several GPUs.

HAT Super Resolution is a local AI image upscaling project for developers, researchers, and people working with low-resolution images. It provides pretrained Hybrid Attention Transformer models and code for training your own.

MedSAM is an open-source model for medical image segmentation that runs locally, aimed at researchers and developers working with medical scans. It builds on Meta's Segment Anything model and uses bounding boxes to identify the region you want to segment. The project uses the Apache 2.0 license and supports macOS.

verl is a Python library for researchers and engineers training large language models on their own hardware. It implements HybridFlow for reinforcement learning after initial training, with support for supervised fine-tuning too.

StyleGAN3 is NVIDIA's local AI image generation and training software for researchers and developers. Its architecture addresses generated detail that stays fixed to pixel positions instead of following the depicted objects as they move.

scikit-learn is a Python machine learning library for people building predictive models and analyzing data. It combines algorithms with data preparation and evaluation tools so you can compare approaches within one library.

Optuna is a Python framework for machine learning developers and researchers who want to automate the search for model settings. Its search spaces can change with the choices made during a trial, so it can compare model architectures or classifiers as well as tune individual parameters. It's open source under the MIT license and can run on your own hardware as part of a Python workload.

Dask-ML is a Python machine learning library for people whose datasets exceed RAM or whose training and evaluation jobs take too long. It uses Dask to distribute work across a cluster of machines, while keeping interfaces familiar to users of Scikit-Learn, NumPy, and pandas. It's open source under the BSD-3-Clause license.

SLEAP helps behavioral researchers measure animal movement in recorded video. It trains models to estimate poses for single animals or groups, with human review during dataset creation and prediction checking.

MLX VLM is a Python package for developers and researchers who want to run multimodal models on their own Mac. It supports local inference and fine-tuning through MLX.

FATE is a self-hosted federated learning framework for enterprises and institutions that want to train models together while protecting data privacy. It supports collaboration across participating organizations and uses homomorphic encryption and multi-party computation (MPC) for secure computation.

Torch-TensorRT compiles PyTorch models with TensorRT to reduce inference latency on NVIDIA hardware. It's for developers running models on their own machines or servers who want faster execution while keeping their PyTorch workflow. The project is open source under the BSD 3-Clause license.

dstack is a self-hosted AI compute orchestration tool for teams running development, training and inference across cloud providers and their own servers. It gives those workloads a shared interface across Kubernetes, VMs and bare-metal clusters.

HunyuanDiT is a local text-to-image model for people generating images from Chinese or English prompts. It handles detailed descriptions and Chinese cultural references, including poetry, idioms, food and architecture.

mergekit combines pretrained language models into one model for developers and researchers who want to combine specialized skills or balance model behaviors. It works directly with model weights, without additional training or access to the original training data.