Model Development Open-Source Projects

Showing 40 indexable projects from the current snapshot.

• huggingface/transformers

  Transformers centralizes state-of-the-art model definitions for text, vision, audio, video, and multimodal tasks, providing a unified API for inference and training. It targets AI engineers, researchers, and developers who need to share tr…

• pytorch/pytorch

  A Python library providing GPU-accelerated tensor computation and a dynamic neural network framework built on a tape-based autograd system. It is designed for researchers and developers who require flexibility and speed for deep learning m…

• huggingface/diffusers

  Diffusers provides modular inference pipelines, interchangeable schedulers, and pretrained building-block models for diffusion-based image, audio, and video generation in PyTorch. It is intended for developers and researchers who need to r…

• karpathy/autoresearch

  This project provides an autonomous research loop where an AI agent modifies training code, runs short experiments, evaluates results, and iterates automatically to improve a small LLM. It requires a single NVIDIA GPU and uses a fixed 5-mi…

• mlabonne/llm-course

  A free, structured educational course covering Large Language Model theory and practical application from fundamentals to deployment, organized into three progressive sections with interactive Colab notebooks. It is intended for researcher…

• d2l-ai/d2l-zh

  A freely available open-source deep learning book that combines mathematical theory with runnable code. Readers can modify and execute Python/Jupyter examples directly while studying the underlying formulas.

• hiyouga/LlamaFactory

  A framework that provides zero-code fine-tuning for over 100 large language models and vision-language models through command-line and web interfaces. It enables continuous pre-training, supervised fine-tuning, and preference alignment usi…

• unslothai/unsloth

  Unsloth provides a local web UI and Python library for running and training large language models with reduced VRAM usage and faster speeds. It supports text, audio, embedding, and vision models, and operates via local or self-hosted execu…

• keras-team/keras

  Keras 3 is a high-level deep learning framework that lets you build, train, and run models across JAX, TensorFlow, PyTorch, and OpenVINO. It is designed for developers and researchers who need to accelerate model development and scale trai…

• jingyaogong/minimind

  An open-source educational project that enables individuals to train and understand a 64M-parameter large language model from scratch in approximately 2 hours. It provides a full PyTorch-based training pipeline, model architecture code, op…

• JuliaLang/julia

  Julia is a high-level, dynamic programming language designed for technical computing, combining ease of use with high performance for numerical and scientific tasks. It runs on cross-platform systems via local binaries or source builds.

• GokuMohandas/Made-With-ML

  A course with accompanying code that teaches developers to design, develop, deploy, and iterate on production-grade machine learning applications. It covers the combination of machine learning and software engineering practices needed to b…

• roboflow/supervision

  Supervision provides reusable Python building blocks for computer vision workflows, including model-agnostic connectors for detection and segmentation results and customizable visualization annotators. It is intended for developers who nee…

• Lordog/dive-into-llms

  A free, hands-on LLM programming tutorial series derived from Shanghai Jiao Tong University courses, covering topics from fine-tuning to AI agents. It provides slides, guides, and Jupyter notebooks for learners who want practical experienc…

• aymericdamien/TensorFlow-Examples

  This repository provides concise tutorials and source code for learning TensorFlow v1 and v2, covering basic models, neural networks, data management, and hardware. It is designed for developers and educators who need clear, executable exa…

• ray-project/ray

  Ray provides a unified framework for scaling Python and AI applications from a laptop to a cluster. It offers distributed abstractions for tasks and actors, alongside libraries for data processing, training, tuning, reinforcement learning,…

• deepspeedai/DeepSpeed

  DeepSpeed is a PyTorch-integrated library that applies system innovations such as ZeRO, ZeRO-Infinity, 3D-Parallelism, and Ulysses Sequence Parallelism to make large-scale deep learning training and inference more efficient and effective.

• hpcaitech/ColossalAI

  ColossalAI provides parallel components that let developers write distributed deep learning models for training and inference with reduced hardware constraints. It targets AI engineers and researchers working on Linux systems with NVIDIA G…

• shiyu-coder/Kronos

  A pre-trained foundation model specialized for financial candlestick (K-line) sequences, supporting forecasting and fine-tuning for quantitative data analysis. It is designed for data teams and researchers who need to process high-noise fi…

• lllyasviel/ControlNet

  lllyasviel/controlnet is a neural network structure for controlling text-to-image diffusion models by adding extra spatial conditions. It learns conditional control on small datasets using zero convolution layers, without destroying the or…

• yunjey/pytorch-tutorial

  This project provides concise PyTorch tutorial code for learning deep learning, covering basics through advanced models. It is designed for educational purposes and local execution rather than production deployment.

• sgl-project/sglang

  SGLang is a serving framework for large language and multimodal models, providing runtime features such as RadixAttention and tensor parallelism across diverse hardware. It is designed for teams that need low-latency, high-throughput infer…

• d2l-ai/d2l-en

  An open-source interactive deep learning book, drafted entirely in Jupyter notebooks, that integrates exposition figures, math, and self-contained runnable code. The resource is freely accessible online and adopted at 500 universities acro…

• liguodongiot/llm-action

  Shares technical principles and practical experience for large-language-model engineering through tutorials covering full fine-tuning, parameter-efficient fine-tuning, and RLHF. Several tutorials include accompanying code links, providing…

• AiHubCN/Awesome-Chinese-LLM

  A curated catalog of over 100 open-source Chinese large language model resources, aggregating base models, domain-specific fine-tunes, datasets, and tutorials. It serves as a reference index for developers and researchers to locate smaller…

• recommenders-team/recommenders

  A repository providing Python utilities, examples, and best practices for prototyping, evaluating, and operationalizing recommendation systems. It covers classical and deep learning algorithms and supports a workflow from data preparation…

• huggingface/peft

  This project enables adaptation of large pretrained models by fine-tuning only a small fraction of extra parameters, reducing computational and storage costs while maintaining performance comparable to fully fine-tuned models. It is design…

• AI4Finance-Foundation/FinGPT

  FinGPT provides open-source financial large language models and benchmarks intended for low-cost fine-tuning on financial tasks. It targets AI engineers and researchers who can operate Python tooling and, when running models locally, suppl…

• kvcache-ai/ktransformers

  ktransformers enables high-performance CPU-GPU heterogeneous computing for large language models, including MoE architectures, reducing hardware requirements for inference and fine-tuning. It provides kt-kernel serving with AMX/AVX acceler…

• ymcui/Chinese-LLaMA-Alpaca

  chinese-llama-alpaca provides open-source Chinese LLaMA and Alpaca models, training scripts, and local deployment tools to improve Chinese semantic understanding and instruction following.

• lightgbm-org/LightGBM

  lightgbm-org/lightgbm is a fast, distributed, high-performance gradient boosting framework that uses tree-based learning algorithms. It delivers faster training speed, lower memory usage, and better accuracy when working with large-scale d…

• stas00/ml-engineering

  An open collection of methodologies, scripts, and copy-n-paste instructions for training, fine-tuning, inferring, and debugging large language models and multi-modal models. It is intended for practitioners who already possess technical kn…

• pytorch/vision

  PyTorch Vision provides the core building blocks—datasets, transforms, and model architectures—needed for computer vision tasks in PyTorch. It is a comprehensive library offering popular public datasets, common image transformations, and p…

• NVIDIA-NeMo/Speech

  A PyTorch-based generative AI framework for researchers and developers to build, customize, and deploy Automatic Speech Recognition, Text-to-Speech, and Speech Large Language Models. The project requires a pre-configured Python environment…

• camel-ai/camel

  An open-source framework for building, operating, and studying multi-agent systems at large scale, supporting up to one million agents. It provides role-playing task automation, synthetic data generation, and retrieval-augmented generation…

• argoproj/argo-workflows

  argo-workflows is an open source, container-native workflow engine for running compute-intensive machine learning and data processing jobs on Kubernetes. It models multi-step workflows as sequences of tasks or directed acyclic graphs (DAGs…

• ddbourgin/numpy-ml

  numpy-ml is a collection of common machine learning algorithms implemented exclusively in NumPy. It is designed to be a transparent reference and starting point for educational purposes and rapid prototyping rather than optimized productio…

• treeverse/dvc

  DVC provides Git-like version control for data and models, lightweight ML pipelines, and local experiment tracking without requiring servers. It addresses the challenge of versioning data, models, and ML experiments while ensuring reproduc…

• tracel-ai/burn

  burn is a Rust deep learning framework that unifies training and inference through a single API where the exact code used for training runs in production. It targets developers who need to move models from experimentation to deployment wit…

• NVIDIA/DeepLearningExamples

  A collection of reproducible deep learning training and deployment scripts optimized for NVIDIA Volta, Turing, and Ampere GPUs using the CUDA-X software stack. It targets AI engineers and researchers who need to run or customize model trai…