Awesome Open Source AI
A curated index of open-source AI projects, models, tools, and infrastructure.
Core open-source AI projects
Core libraries and frameworks used to build, train, and run AI and machine learning systems.
PyTorchDynamic computation graphs, Pythonic API, dominant in research and production. The current standard for most frontier AI work. 100k–Python
TensorFlowEnd-to-end platform with excellent production deployment, TPU support, and large-scale serving tools. 195kApache-2.0C++
JAXHigh-performance numerical computing with composable transformations (JIT, vmap, grad). Rising favorite for research and scientific ML. + Flax 36kApache-2.0Python
dm-haikuJAX-based neural network library from Google DeepMind. Elegant functional API with state management, widely used in DeepMind's research. Apache 2.0 licensed. 3.2kApache-2.0Python
EquinoxElegant easy-to-use neural networks and scientific computing in JAX. Callable PyTrees with filtered transformations, seamless interoperability with the JAX ecosystem. Apache 2.0 licensed. 2.9kApache-2.0PythonDiffraxNumerical differential equation solvers in JAX. Autodifferentiable and GPU-capable ODE/SDE/CDE solvers for scientific machine learning and neural differential equations. Apache 2.0 licensed. 2kApache-2.0Python
vit-pytorchComprehensive Vision Transformer (ViT) implementations in PyTorch. Reference implementations of all major vision transformer variants including ViT, DeiT, Swin, and more. MIT licensed. 25kMITPython
NumPyroProbabilistic programming with NumPy powered by JAX for autograd and JIT compilation. Bayesian modeling and inference at scale. 2.7kApache-2.0Python
KerasHigh-level, beginner-friendly API that now runs on multiple backends (TensorFlow, JAX, PyTorch). Perfect for rapid experimentation. 64kApache-2.0Python
tinygradMinimalist deep learning framework with tiny code footprint. The "you like PyTorch? you like micrograd? you love tinygrad!" philosophy - simple yet powerful. 33kMITPython
PaddlePaddleIndustrial deep learning platform from Baidu serving 23+ million developers and 760,000+ companies. China's first independent R&D framework with advanced distributed training and deployment capabilities. 24kApache-2.0C++
PyTorch GeometricLibrary for deep learning on irregular input data such as graphs, point clouds, and manifolds. Part of the PyTorch ecosystem. 24kMITPython
timm (PyTorch Image Models)The largest collection of PyTorch image encoders and backbones. 900+ pretrained models including ResNet, EfficientNet, Vision Transformer, ConvNeXt, and more with training and inference scripts. Apache 2.0 licensed. 37kApache-2.0Python
TritonLanguage and compiler for writing highly efficient custom deep-learning primitives. Powers kernel optimizations in PyTorch, JAX, and other frameworks. MIT licensed. 19kMITMLIR
oneDNNoneAPI Deep Neural Network Library. Cross-platform performance library of basic building blocks for deep learning, optimized for Intel CPUs, GPUs, and Arm architectures. Apache 2.0 licensed. 4kApache-2.0C++
Flux.jl100% pure-Julia ML stack with lightweight abstractions on top of native GPU and AD support. Elegant, hackable, and fully integrated with Julia's scientific computing ecosystem. 4.7k–Julia
spaCy (Explosion AI)Industrial-strength natural language processing with 75+ languages, transformer pipelines, and production-grade NER, parsing, and text classification. 34kMITPythonTransformers (Hugging Face)The de facto standard library for pretrained NLP models. 1M+ models, 250,000+ downloads/day. BERT, GPT, Llama, Qwen, and hundreds more. 161kApache-2.0Pythontokenizers (Hugging Face)Fast state-of-the-art tokenizers for training and inference. 11kApache-2.0Rust
PolarsBlazing-fast DataFrame library (Rust backend) - modern alternative to Pandas for large-scale workloads. 39kMITRust
cuDFGPU DataFrame library from RAPIDS. Accelerates Pandas workflows on NVIDIA GPUs with zero code changes using cuDF.pandas accelerator mode. 9.7kApache-2.0C++
ModinParallel Pandas DataFrames. Scale Pandas workflows by changing a single line of code - distributes data and computation automatically. 10kApache-2.0Python
DaskParallel computing for big data - scales Pandas/NumPy/scikit-learn to clusters. 14kBSD-3-ClausePython
DataFlowLLM-ready data preparation system for turning raw PDFs, conversations, code, databases, and other sources into SFT, QA, and RAG-ready datasets. 6.2kApache-2.0Python
SciPyScientific computing algorithms (optimization, linear algebra, statistics, signal processing). 15kBSD-3-ClausePython
NetworkXCreation, manipulation, and study of complex networks. The foundational graph analysis library for Python data science. 17k–PythoncuGraphGPU graph analytics library with NetworkX-compatible API. 10-100x faster than CPU for large-scale graph algorithms. Apache 2.0 licensed. 2.2kApache-2.0Cuda
VaexOut-of-Core hybrid Apache Arrow/NumPy DataFrame for Python. Visualize and explore billion-row datasets at millions of rows per second. MIT licensed. 8.5kMITPython
DatashaderHigh-performance large data visualization. Renders billions of points interactively without aggregation artifacts. BSD-3-Clause licensed. 3.5kBSD-3-ClausePython
ZarrChunked, compressed, N-dimensional array storage. Scalable tensor data format optimized for cloud and parallel computing. MIT licensed. 2kMITPython
NVIDIA DALIGPU-accelerated data loading and augmentation library with highly optimized building blocks for deep learning applications. Apache 2.0 licensed. 5.7kApache-2.0C++
NarwhalsLightweight compatibility layer between DataFrame libraries. Write Polars-like code that works seamlessly across Pandas, Polars, cuDF, Modin, and more. MIT licensed. 1.6kMITPython
IbisPortable Python dataframe library with 20+ backends. Write Pandas-like code that runs locally with DuckDB or scales to production databases (BigQuery, Snowflake, PostgreSQL) by changing one line. Apache 2.0 licensed. 6.6kApache-2.0Python
skrubMachine learning with dataframes for dirty categorical data. Preprocessing and feature engineering for heterogeneous data with seamless Pandas/Polars integration. BSD-3-Clause licensed. 1.6kBSD-3-ClausePython
OxenLightning fast data version control for machine learning. Optimized for large datasets with efficient diffing, branching, and collaboration. Apache 2.0 licensed. 1.1kApache-2.0Rust
PanderaStatistical data testing and validation for dataframes. Pydantic-like API for Pandas, Polars, and other dataframe libraries with type hints and lazy validation. MIT licensed. 4.4kMITPython
SnorkelSystem for quickly generating training data with weak supervision. Programmatically label, build, and manage training data using labeling functions and probabilistic consensus models. Powers Snorkel Flow and used by Google, Apple, and Intel. Apache 2.0 licensed. 6kApache-2.0Python
DuckDBHigh-performance analytical in-process SQL database system. Fast, reliable, portable, and easy to use with rich SQL dialect support. Perfect for data processing and analytics workloads. MIT licensed. 39kMITC++
FiftyOneVisual AI development toolkit for visualizing, labeling, and evaluating visual datasets and models. Supercharges computer vision workflows with dataset exploration and model analysis. Apache 2.0 licensed. 11kApache-2.0Python
Label StudioMulti-type data labeling and annotation tool with standardized output format. Configurable interface for images, text, audio, video, and time series with ML-assisted labeling. Apache 2.0 licensed. 28kApache-2.0TypeScript
Delta LakeOpen-source storage framework enabling Lakehouse architecture with ACID transactions, scalable metadata handling, and unified batch/streaming processing. Apache 2.0 licensed. 8.8kApache-2.0Scala
Apache IcebergHigh-performance open table format for huge analytic tables. Brings SQL table reliability to big data with time travel, hidden partitioning, and schema evolution. Works with Spark, Trino, Flink, Presto, Hive and Impala. Apache 2.0 licensed. 8.9kApache-2.0JavaApache HudiOpen data lakehouse platform for ingesting, indexing, storing, serving, transforming and managing data across cloud environments. Supports upserts, deletes and incremental processing on big data with built-in ingestion tools for Spark and Flink. Apache 2.0 licensed. 6.2kApache-2.0Java
lakeFSData version control for your data lake that transforms object storage into Git-like repositories. Enables atomic, versioned data lake operations with branching, committing, and merging for data pipelines. Apache 2.0 licensed. 5.4kApache-2.0GoApache AirflowPlatform to programmatically author, schedule, and monitor workflows. Industry-standard orchestration for data pipelines and ML workflows with 500+ integrations. Apache 2.0 licensed. 46kApache-2.0PythonApache SparkUnified analytics engine for large-scale data processing. In-memory cluster computing with high-level APIs in Python, Scala, Java, and R. Powers MLlib for distributed machine learning and Structured Streaming for real-time data. Apache 2.0 licensed. 43kApache-2.0ScalaApache FlinkStream processing framework with powerful batch and streaming capabilities. High-throughput, low-latency runtime with exactly-once processing guarantees. Ideal for real-time AI inference pipelines and event-driven ML applications. Apache 2.0 licensed. 26kApache-2.0JavaApache BeamUnified programming model for batch and streaming data processing. Write pipelines once, run anywhere on Flink, Spark, or Google Cloud Dataflow. Portable, extensible, and enterprise-ready for AI data pipelines. Apache 2.0 licensed. 8.6kApache-2.0Java
ScrapyFast, high-level web crawling and scraping framework for Python. Extract structured data from websites at scale with built-in support for handling common challenges like pagination, cookies, and concurrent requests. BSD-3-Clause licensed. 62kBSD-3-ClausePython
TemporalDurable execution platform for reliable workflow orchestration. Build resilient data pipelines and ML workflows that survive failures and continue execution exactly where they left off. MIT licensed. 21kMITGo
LuigiPython module for building complex pipelines of batch jobs. Handles dependency resolution, workflow management, visualization, and Hadoop integration. Built at Spotify and battle-tested in production. Apache 2.0 licensed. 19kApache-2.0Python
Mage.aiModern open-source data pipeline tool for integrating and transforming data. AI-native ETL/ELT platform with 100+ integrations, real-time monitoring, and collaborative features. Apache 2.0 licensed. 8.7kApache-2.0PythonHamiltonDeclarative dataflow framework for building testable, modular, self-documenting data pipelines. Encode lineage and metadata directly in Python functions. Originally from Stitch Fix, now Apache incubating. Apache 2.0 licensed. 2.5kApache-2.0Jupyter Notebook
D-TaleVisualizer for Pandas data structures with a Flask back-end and React front-end. Interactive data exploration with charting, filtering, and code export. LGPL-2.1 licensed. 5.2kLGPL-2.1TypeScript
SweetvizBeautiful, high-density visualizations for exploratory data analysis in two lines of code. Self-contained HTML reports for dataset comparison and target analysis. MIT licensed. 3.1kMITPython
TextAttackPython framework for adversarial attacks, data augmentation, and model training in NLP. Augment datasets to increase model robustness and generate adversarial examples. MIT licensed. 3.4kMITPython
scikit-learnIndustry-standard library for traditional machine learning (classification, regression, clustering, pipelines). 66kBSD-3-ClausePython
XGBoostScalable, high-performance gradient boosting library. Still dominates Kaggle and tabular competitions. 28kApache-2.0C++
LightGBMMicrosoft's ultra-fast gradient boosting framework, optimized for speed and memory. 18kMITC++
CatBoostGradient boosting that handles categorical features natively with great out-of-the-box performance. 9kApache-2.0C++
sktimeUnified framework for machine learning with time series. scikit-learn compatible API for forecasting, classification, clustering, and anomaly detection. 9.8kBSD-3-ClausePython
StatsForecastLightning-fast statistical forecasting with ARIMA, ETS, CES, and Theta models. Optimized for high-performance time series workloads. 4.8kApache-2.0PythonMLForecastScalable machine learning for time series forecasting. Train any sklearn-compatible model on millions of time series with efficient feature engineering. Apache 2.0 licensed. 1.2kApache-2.0PythoncuMLGPU-accelerated machine learning algorithms with scikit-learn compatible API. 10-50x faster than CPU implementations for large datasets. Apache 2.0 licensed. 5.2kApache-2.0Python
SynapseMLDistributed machine learning on Apache Spark. Scalable, composable APIs for text analytics, vision, anomaly detection with seamless Python/Scala/R/.NET integration. MIT licensed. 5.2kMITScala
DataHubThe 1 open-source metadata platform for data and AI. Data discovery, governance, and observability with 80+ connectors, column-level lineage, and AI assistant integration. Originally built at LinkedIn. Apache 2.0 licensed. 12kApache-2.0Python
dbt-coreTransform data using software engineering best practices. The industry-standard framework for analytics engineering with 15M+ monthly downloads. Enables version control, testing, and documentation for SQL transformations. Apache 2.0 licensed. 13kApache-2.0Rust
DeequLibrary built on top of Apache Spark for defining "unit tests for data". Measures data quality in large datasets with constraint verification, anomaly detection, and incremental validation. Used at Amazon for production data quality. Apache 2.0 licensed. 3.6kApache-2.0Scala
Great ExpectationsAlways know what to expect from your data. Data validation, profiling, and documentation for data pipelines. Apache 2.0 licensed. 12kApache-2.0Python
ydata-profilingOne line of code for comprehensive data quality profiling and exploratory data analysis. Generates detailed reports for Pandas and Spark DataFrames including statistics, correlations, missing values, and data quality alerts. MIT licensed. 14kMITPython
OptunaModern, define-by-run hyperparameter optimization with pruning and visualizations. Extremely popular in 2026. 14kMITPython
AutoGluonAWS AutoML toolkit for tabular, image, text, and multimodal data - state-of-the-art with almost zero code. 10kApache-2.0Python
StreamlitThe fastest way to build and share data apps. Transform Python scripts into beautiful web applications with minimal code. Widely used for ML model demos, data visualization, and internal tools. 45kApache-2.0Python
Hugging Face AccelerateSimple API to make training scripts run on any hardware (multi-GPU, TPU, mixed precision) with minimal code changes. 9.7kApache-2.0Python
DeepSpeedMicrosoft's deep learning optimization library for extreme-scale training (ZeRO, offloading, MoE). 42kApache-2.0Python
FlashAttentionFast exact attention kernels that reduce memory usage and accelerate transformer training and inference. 24kBSD-3-ClausePython
PyTorch LightningHigh-level wrapper for PyTorch that removes boilerplate and adds best practices. 31kApache-2.0Python
fastaiDeep learning library providing practitioners with high-level components for state-of-the-art results. Built on PyTorch with a focus on usability and transfer learning. Apache 2.0 licensed. 28kApache-2.0Jupyter NotebookPyTorch IgniteHigh-level library for training and evaluating neural networks in PyTorch with an engine, events & handlers system for maximum flexibility. BSD-3-Clause licensed. 4.8kBSD-3-ClausePython
einopsFlexible, powerful tensor operations for readable and reliable code. Supports PyTorch, JAX, TensorFlow, NumPy, MLX. 9.5kMITPython
safetensorsSimple, safe way to store and distribute tensors. Fast, secure alternative to pickle for model serialization. 3.8kApache-2.0RusttorchmetricsMachine learning metrics for distributed, scalable PyTorch applications. 80+ metrics with built-in distributed synchronization. 2.4kApache-2.0PythontorchaoPyTorch native quantization and sparsity for training and inference. Drop-in optimizations for production deployment. 2.8k–Python
SHAPGame theoretic approach to explain the output of any machine learning model. Industry standard for model interpretability. 26kMITJupyter Notebook
skorchscikit-learn compatible neural network library that wraps PyTorch. Seamlessly integrate PyTorch models with scikit-learn pipelines, grid search, and cross-validation. 6.2kBSD-3-ClauseJupyter NotebookModel open-source AI projects
Canonical model-family repositories with useful code, recipes, evaluation tools, or engineering context. This is not a complete model leaderboard; use Hugging Face and model hubs for up-to-date weight discovery.
RWKVAttention-free language model architecture with linear-time inference, training code, inference examples, and an active open-source ecosystem. 15kApache-2.0Python
QwenCanonical open model family from Alibaba with model cards, inference examples, fine-tuning guidance, and ecosystem links. 27k–PythonQwen3-VLVision-language branch of the Qwen3 model stack with multimodal checkpoints, training recipes, and reference tooling. 19kApache-2.0Jupyter Notebook
MiniCPMCompact open model family with practical code, deployment notes, and active edge/on-device focus. 9.4kApache-2.0Jupyter Notebook
Llama ModelsMeta's canonical repository for Llama model documentation, examples, safety materials, and integration guidance. 7.6k–Python
GPT-OSSOpenAI open-weight model repository with inference examples, recipes, and deployment guidance. 20kApache-2.0Python
MambaState Space Model implementation with pretrained checkpoints, architecture code, and research tooling for efficient long-sequence modeling. 18kApache-2.0Python
openai/CLIPCanonical OpenAI contrastive vision-language model codebase with pretrained checkpoints and practical reference implementation for image-text retrieval and classification. 34kMITJupyter Notebook
OpenCLIPOpen implementation of CLIP with training code, pretrained models, and zero-shot evaluation tooling. 14k–PythonOmniParserVision-based GUI parsing model and tooling for computer-use agents. 25kCC-BY-4.0Jupyter NotebookMiniCPM-VCompact vision-language model family with edge-focused deployment examples and strong OCR-oriented use cases. 26kApache-2.0Python
EagleNVIDIA multimodal model codebase with open checkpoints and reusable research materials for vision-language and video-language tasks. 2.2kApache-2.0Python
MoondreamSmall vision-language model with practical inference examples for edge and real-time image understanding. 9.7kApache-2.0PythonVILANVIDIA vision-language model family with training, evaluation, and deployment materials across edge and datacenter settings. 3.8kApache-2.0Python
WhisperCanonical open speech-to-text model codebase with widespread ecosystem support and many downstream implementations. 101kMITPython
FunASRSpeech recognition toolkit with pretrained models, streaming support, diarization, VAD, and production-oriented examples. 17kMITPython
NVIDIA NeMoScalable framework and model codebase for speech, language, and multimodal AI with recipes and deployment guidance. 17kApache-2.0Python
Sherpa-ONNXComplete speech toolkit with ASR, TTS, diarization, source separation, and VAD across embedded and edge environments via ONNX Runtime. 13kApache-2.0C++
MOSS-TTSOpen speech and sound generation family focused on expressive, long-form text-to-speech with streaming and multi-speaker support. 3.1kApache-2.0PythonVoxCPMOpen-sourced tokenizer-free multilingual speech synthesis model with high-quality TTS and style transfer workflows. 26kApache-2.0Python
Serving open-source AI projects
Inference runtimes, serving systems, and optimization tools for running models locally or in production.
llama.cppPure C/C++ inference engine with GGUF format support. The gold standard for CPU/GPU/Apple Silicon on-device running. Includes llama-server for OpenAI-compatible API. Now at 100K+ stars. 115kMITC++
OllamaDead-simple local LLM runner with a one-line install, model registry, and OpenAI-compatible API. 173kMITGoFoundry LocalOpen-source on-device AI platform covering discovery, model running, sandboxed execution, and evaluation of open models. 2.4k–C++
Potato OSLinux distribution for fully local AI inference on Raspberry Pi 5 and 4, optimized for running open models at the edge. 93Apache-2.0Python
MLC-LLMDeployment engine that compiles and runs LLMs across browsers, mobile devices, and local hardware. 23kApache-2.0PythonWebLLMHigh-performance in-browser LLM inference engine. Runs models directly in the browser with WebGPU acceleration. 18kApache-2.0TypeScript
RamaLamaContainer-centric tool for simplifying local AI model serving. Automatically detects GPUs, pulls optimized container images, and runs models securely in rootless containers with enterprise-grade isolation. 2.9kMITPython
LiteRTGoogle's production-ready on-device ML and GenAI deployment framework. Supports Android, iOS, Web, Desktop, and IoT targets with GPU/NPU acceleration via a unified edge-first runtime. Apache 2.0 licensed. 2.5kApache-2.0C++LiteRT-LMProduction-ready runtime for deploying LLMs on edge devices with low-latency inference and optimized hardware paths for mobile and embedded platforms. 5.4kApache-2.0C++
exoRun frontier AI locally by connecting all your devices into an AI cluster. Features automatic device discovery, RDMA over Thunderbolt for 99% latency reduction, topology-aware auto parallel, and tensor parallelism. Uses MLX backend for distributed inference across Apple Silicon devices. Apache 2.0 licensed. 45kApache-2.0Python
llm-dKubernetes-native distributed LLM inference framework. Donated to CNCF by RedHat, Google, and IBM. Intelligent scheduling, KV-cache optimization, and state-of-the-art performance across accelerators. 3.3kApache-2.0Shell
LMDeployToolkit for compressing, deploying, and serving LLMs from OpenMMLab. 4-bit inference with 2.4x higher performance than FP16, distributed multi-model serving across machines. 7.9kApache-2.0Python
vLLMState-of-the-art serving engine with PagedAttention and continuous batching. Currently the fastest production-grade LLM server. 82kApache-2.0PythonvLLM-OmniMulti-modal inference stack extending vLLM for image, audio, and video generation workloads with a unified serving interface. Apache 2.0 licensed. 5kApache-2.0Python
LMCacheSupercharge LLM inference with the fastest KV Cache layer. 3-10x delay savings and GPU cycle reduction for multi-round QA and RAG. Integrates seamlessly with vLLM for distributed, high-throughput deployments. Apache 2.0 licensed. 8.4kApache-2.0PythonvLLM Production StackKubernetes-native production stack for vLLM inference. Automated deployment, autoscaling, and monitoring for enterprise-grade LLM serving. Built by the vLLM team for seamless integration. 2.4kApache-2.0Python
Open Model Engine (OME)Kubernetes operator for LLM serving with GPU scheduling and model lifecycle management across vLLM, SGLang, and TensorRT-LLM. 461Apache-2.0Go
nano-vLLMMinimalist vLLM implementation in ~1,200 lines of Python. Educational yet performant with prefix caching, tensor parallelism, and CUDA graph acceleration. Comparable inference speeds to full vLLM. MIT licensed. 14kMITPython
SGLangNext-gen serving framework with RadixAttention. Powers xAI's production workloads at 100K+ GPUs scale. 29kApache-2.0Python
Aphrodite EnginevLLM fork optimized for role-play and creative writing. Supports extensive quantization methods (AQLM, AWQ, GPTQ, GGUF, FP8) and modern samplers. Active development with multi-LoRA and speculative decoding support. 1.8kAGPL-3.0C++AIBrixCost-efficient and pluggable infrastructure components for GenAI inference. Kubernetes-native control plane for vLLM with distributed KV cache, heterogeneous GPU serving, and intelligent routing. Apache 2.0 licensed. 4.8kApache-2.0Go
Triton Inference ServerNVIDIA's production-grade open-source inference serving software. Supports multiple frameworks (TensorRT, PyTorch, ONNX) with optimized cloud and edge deployment. 11kBSD-3-ClausePython
mistral.rsFast, flexible Rust-native LLM inference engine built on Candle. Supports text, vision, audio, image generation, and embeddings with hardware-aware auto-tuning. 7.2kMITRust
KTransformersFlexible framework for heterogeneous CPU-GPU LLM inference and fine-tuning. Enables running large MoE models by offloading experts to CPU with BF16/FP8 precision support. 17kApache-2.0Python
llamafileMozilla's single-file distributable LLM solution. Bundle model weights, inference engine, and runtime into one portable executable that runs on six OSes without installation. 25k–C++
XinferenceUnified, production-ready inference API for LLMs, speech, and multimodal models. Drop-in GPT replacement with single-line code changes. Supports thousands of models with auto-batching and distributed inference. 9.3kApache-2.0Python
RTP-LLM (Alibaba)Alibaba's high-performance LLM inference acceleration engine. Powers production LLM services across Taobao, Tmall, and Alibaba's international AI platform. Supports PagedAttention, FlashAttention, FlashDecoding, INT8/INT4 quantization, and heterogeneous hardware (GPU/ARM CPU/Intel). Apache 2.0 licensed. 1.2kApache-2.0CudaLitServe (Lightning AI)Minimal Python framework for building custom AI inference servers with full control over logic, batching, and scaling. 2x faster than FastAPI with built-in batching, streaming, and multi-GPU autoscaling. Apache 2.0 licensed. 3.9kApache-2.0Python
LightLLMPure Python-based LLM inference and serving framework with lightweight design, easy extensibility, and high-speed performance. Integrates optimizations from FasterTransformer, TGI, vLLM, and SGLang. 4.1kApache-2.0Python
TabbyAPIFastAPI-based API server for ExLlamaV2/V3 backends. OpenAI-compatible API with support for model loading/unloading, embeddings, speculative decoding, multi-LoRA, and streaming. 1.2kAGPL-3.0Python
GPUStackGPU cluster manager that orchestrates inference engines like vLLM and SGLang. Automated engine selection, parameter optimization, and distributed multi-GPU deployment for high-performance AI workloads. 5.1kApache-2.0Python
One-APILLM API management and key redistribution system. Unifies multiple providers (OpenAI, Anthropic, Azure, etc.) under a single OpenAI-compatible API with built-in rate limiting, quota management, and cost tracking. MIT licensed. 35kMITJavaScript
OpenLLM (BentoML)Production-grade platform for running any open-source LLMs as OpenAI-compatible API endpoints. Supports 50+ models with built-in streaming, batching, and auto-acceleration. Apache 2.0 licensed. 12kApache-2.0Python
Higress (Alibaba)AI-native API gateway born from Alibaba's internal infrastructure with 2+ years of production validation. Provides unified LLM API and MCP (Model Context Protocol) management with enterprise-grade 99.99% availability. Apache 2.0 licensed. 8.6kApache-2.0Go
NVIDIA DynamoDatacenter-scale distributed inference serving framework from NVIDIA. Orchestration layer above vLLM/SGLang/TensorRT-LLM with disaggregated serving, KV-aware routing, and automatic scaling. Built in Rust with Python extensibility. Apache 2.0 licensed. 7.2k–RustMicrosoft BitNetOfficial inference framework for 1-bit LLMs (BitNet b1.58). Enables running large models on CPU with minimal memory footprint. Features custom kernels for ternary weight quantization and efficient matmul operations. MIT licensed. 39kMITPython
DeepEPEfficient expert-parallel communication library for large MoE models, improving throughput in distributed inference and training. 9.7kMITCudaDeepGEMMCUDA FP8/FMA GEMM kernels for efficient LLM inference and training at reduced precision. 7.3kMITCuda
AirLLMSingle-GPU 70B inference stack with strong memory/performance optimizations for local deployment on commodity hardware. 20kApache-2.0Jupyter Notebook
ThunderKittensHigh-performance GPU kernel primitives for fast attention and matmul workflows used by LLM stacks. 3.4kMITCuda
Mirage Persistent KernelCompiler that fuses model execution into a single mega-kernel for tighter performance. 2.3kApache-2.0Cuda
tt-metalOperator and kernel toolkit for efficient LLM inference and low-level optimization on Tenstorrent hardware. 1.5kApache-2.0C++
CTranslate2Fast inference engine for Transformer models supporting OpenNMT and Hugging Face models. Optimized for CPU and GPU with batching, quantization (INT8/FP16), and dynamic memory management. Powers faster-whisper and other production deployments. MIT licensed. 4.5kMITC++
llama-swapIntelligent model swapping proxy for llama.cpp. Enables seamless hot-swapping between different GGUF models without restarting the server, with automatic model loading/unloading and OpenAI-compatible API. MIT licensed. 4.4kMITGo
optillmOptimizing inference proxy for LLMs with load balancing, failover, and request routing across multiple providers and models. Improves reliability and performance for production deployments. Apache 2.0 licensed. 4.1kApache-2.0Python
FugusashiOpen-source intelligent model router with CMA-ES evolved routing weights, federated learning for privacy-preserving collaborative routing, and human-interpretable explanations for every decision. OpenAI-compatible API with web dashboard. MIT licensed. 5MITPython
mllmFast and lightweight multimodal LLM inference engine for mobile and edge devices. Optimized for running vision-language models on resource-constrained hardware with efficient memory management. MIT licensed. 1.5kMITC++
shimmyPython-free Rust inference server with OpenAI API compatibility. Supports GGUF and SafeTensors formats with hot model swap, auto-discovery, and single binary deployment for zero-dependency inference. Apache 2.0 licensed. 5.3kApache-2.0Rust
PowerInferHigh-speed LLM inference for local deployment on consumer GPUs. Achieves up to 11x speedup over llama.cpp on RTX 4090 by exploiting power-law neuron activation patterns. MIT licensed. 9.5kMITC++
distributed-llamaDistributed LLM inference connecting home devices into a powerful cluster. More devices means faster inference via tensor parallelism over Ethernet. Supports Linux, macOS, Windows, ARM, and x86_64 AVX2 CPUs. MIT licensed. 2.9kMITC++
ik_llama.cppHigh-performance llama.cpp fork with better CPU and hybrid GPU/CPU performance, SOTA quantization types, first-class Bitnet support, and improved DeepSeek performance via MLA, FlashMLA, and fused MoE operations. MIT licensed. 2.7kMITC++
xLLMHigh-performance inference engine optimized for Chinese AI accelerators (Cambricon MLU, Hygon DCU, Huawei Ascend). Features service-engine decoupled architecture with elastic scheduling, PD disaggregation, and global KV cache management. Powers JD.com's core retail businesses. Apache 2.0 licensed. 1.3kApache-2.0C++MooncakeProduction-grade serving platform for Kimi (Moonshot AI). Features distributed KV cache pool with intelligent offloading, prefill/decode disaggregation, and cross-instance KV reuse. Integrated with vLLM, SGLang, and TensorRT-LLM. Apache 2.0 licensed. 5.5kApache-2.0C++
DeepEPCommunication library for efficient expert-parallel training/inference pipelines, reducing MoE cross-device communication overhead. Apache 2.0 licensed. 9.7kMITCudaDeepGEMMClean, high-performance FP8 GEMM kernels with fine-grained scaling for modern inference workloads. Apache 2.0 licensed. 7.3kMITCudaRAFTCUDA-accelerated algorithms and ANN building blocks for high-performance similarity search, clustering, and matrix learning workloads. 1kApache-2.0Cuda
SageAttentionQuantized attention kernels with reported 2-5x speedups versus FlashAttention across text, image, and video models. Apache 2.0 licensed. 3.4kApache-2.0CudaThunderKittensCUDA tile primitives and kernel templates for accelerating transformer attention blocks. MIT licensed. 3.4kMITCuda
Agents open-source AI projects
Frameworks and platforms for building agent-based systems and multi-agent workflows.
AutoGPTThe original autonomous AI agent framework that sparked the agent revolution. Vision of accessible AI for everyone with modular agent architecture, benchmark testing, and forge-based agent building. 183k+ stars. 185k–Python
BabyAGIPioneering task-driven autonomous agent that inspired the AI agent movement. Simple, elegant implementation of an AI agent that creates, prioritizes, and executes tasks autonomously. 22k+ stars. 22k–Python
DSPyFramework for programming language model pipelines with modules, optimizers, and evaluation loops. 35kMITPythonSemantic KernelSDK for building and orchestrating AI agents and workflows across multiple programming languages. 28kMITC#smolagentsLightweight agent framework centered on tool use and code-executing workflows. 28kApache-2.0Python
Neuron AIPHP Agentic Framework for building production-ready AI driven applications. Connect components (LLMs, vector DBs, memory) to agents that can interact with your data. MIT licensed. 2kMITPHP
II-Agent (Intelligent Internet)New open-source framework to build and deploy intelligent agents with support for Claude, Gemini, and OpenAI models. Apache 2.0 licensed. 3.3kApache-2.0Python
Hermes Agent (NousResearch)The agent that grows with you. Autonomous server-side agent with persistent memory that learns and improves over time. 181kMITPython
Strands AgentsModel-driven approach to building AI agents in just a few lines of code. Multi-agent systems, autonomous agents, and streaming support with built-in MCP. Apache 2.0 licensed. 6kApache-2.0Python
AgnoBuild, run, and manage agentic software at scale. High-performance framework for multi-agent systems with memory, knowledge, and tools. 41kApache-2.0Python
UpsonicAgent framework for fintech and banking with built-in MCP support, guardrails, and tool server architecture. 7.9kMITPython
VoltAgentTypeScript-first AI agent engineering platform with memory, RAG, workflows, MCP integration, and voice support. 9.4kMITTypeScript
PocketFlow100-line minimalist LLM framework for building agent workflows. Lightweight, extensible architecture for tool use and autonomous task execution. 11kMITPython
Agent Development Kit (Google)Code-first Python toolkit for building sophisticated AI agents with multi-agent orchestration, built-in evaluation, and flexible deployment. Model-agnostic with tight Google ecosystem integration. Apache 2.0 licensed. 20kApache-2.0Python
PydanticAIType-safe AI agent framework from the creators of Pydantic. Model-agnostic with 20+ providers, built-in observability via Logfire, MCP/A2A protocol support, and YAML/JSON agent definitions. MIT licensed. 18kMITPythonQwen-AgentAgent framework built on Qwen models featuring function calling, MCP support, code interpreter, RAG, and Chrome extension. Powers Qwen Chat with advanced tool use and planning capabilities. Apache 2.0 licensed. 16kApache-2.0Python
GriptapeModular Python framework for AI agents and workflows with chain-of-thought reasoning, tools, and memory. Enforces structures like sequential pipelines and DAG-based workflows for predictable AI systems. Apache 2.0 licensed. 2.5kApache-2.0Python
LangroidHarness LLMs with multi-agent programming. Mature tool calling system based on Pydantic, supports hundreds of LLM providers including OpenAI and local servers. Built for robust agent behavior in real-world use cases. MIT licensed. 4kMITPython
OctomindModel-agnostic AI agent runtime written in Rust with specialist agents, MCP support, multiple provider integrations, and zero-config setup. Apache 2.0 licensed. 66Apache-2.0Rust
MarvinPython framework for structured outputs and agentic AI workflows. Simplifies LLM interactions with type-safe interfaces, automatic schema generation, and built-in observability. From the creators of Prefect. Apache 2.0 licensed. 6.2kApache-2.0PythonBurrApache incubating framework for building stateful AI applications (chatbots, agents, simulations). Monitor, trace, persist, and execute on your own infrastructure with built-in UI and pluggable memory. Apache 2.0 licensed. 2kApache-2.0Python
KaibanJSJavaScript-native framework for building and managing multi-agent systems with a Kanban-inspired approach. Visual task board for AI agents with real-time collaboration features. MIT licensed. 1.4kMITTypeScript
JidoAutonomous agent framework for Elixir. Built for distributed, autonomous behavior and dynamic workflows with actor-model concurrency. Apache 2.0 licensed. 1.7kApache-2.0Elixir
FlueProgrammable TypeScript harness and sandbox agent framework for building autonomous workflows and agents. Apache 2.0 licensed. 5.6kApache-2.0TypeScript
Agent-NativeTypeScript-first framework for building agent-first applications featuring shared database state, real-time multiplayer editing, and action-driven tools. ISC licensed. 1.2k–TypeScript
MetaGPTThe Multi-Agent Framework: First AI Software Company. Assigns different roles to GPTs to form a collaborative software entity. Takes one-line requirements and outputs comprehensive software development artifacts including user stories, competitive analysis, requirements, data structures, APIs, and documents. ICLR 2024 oral presentation (top 1.2%). MIT licensed. 69kMITPythonChatDevMulti-agent software development framework where AI agents collaborate as programmers, designers, and testers to build software. Apache 2.0 licensed. 33kApache-2.0Python
CAMELFirst and best multi-agent framework for building scalable agent systems. Apache 2.0 licensed with extensive tooling for agent communication and task automation. 17kApache-2.0Python
DeepAgentsBatteries-included LangChain agent harness for building and running structured multi-agent workflows with reusable runtime patterns. 24kMITPython
MastraTypeScript-first agent framework with built-in RAG, workflows, tool integrations, observability and observational memory. 25k–TypeScript
Deer-Flow (ByteDance)Open-source long-horizon SuperAgent harness that researches, codes, and creates. Handles tasks from minutes to hours with sandboxes, memories, tools, skills, subagents, and message gateway. 70kMITPythonOpenAI Agents SDKProduction-ready lightweight framework for multi-agent workflows. The evolution of Swarm with enhanced orchestration capabilities and enterprise-grade features. 27kMITPythonSymphonyTurns project work into isolated, autonomous implementation runs. Monitors work boards, spawns agents to handle tasks, and provides proof of work including CI status, PR reviews, and walkthrough videos. Engineering preview for managing work instead of supervising coding agents. Apache 2.0 licensed. 25kApache-2.0Elixir
PaperclipAI agent company and orchestration framework with 55K+ stars. MIT licensed. 69kMITTypeScript
AgentScopeAlibaba's production-ready multi-agent framework with 23K+ stars. Features built-in MCP and A2A support, message hub for flexible orchestration, and AgentScope Runtime for production deployment. 26kApache-2.0Python
mcp-agentBuild effective agents using Model Context Protocol and simple workflow patterns. Handles connection mechanics, LLM integration, and persistent state for production MCP-based agents. MIT licensed. 8.4kApache-2.0PythonMicrosoft Agent FrameworkMicrosoft's official framework combining AutoGen's agent abstractions with Semantic Kernel's enterprise features. Supports Python and .NET with graph-based workflows. 11kMITPython
Agency SwarmReliable multi-agent orchestration framework built on top of the OpenAI Assistants API with organizational structure modeling. 4.4kMITPython
elizaOSAutonomous multi-agent framework for building and deploying AI-powered applications. Features Discord/Telegram/Farcaster connectors, RAG support, and a modern web dashboard. 19kMITTypeScriptOpenManusOpen-source framework for building general AI agents. Modular agent architecture with planning, tool use, and autonomous task execution. 56k+ stars. MIT licensed. 56kMITPython
OpenAgentsAI Agent Networks for Open Collaboration. Platform for building collaborative multi-agent systems with shared knowledge and distributed task execution. Apache 2.0 licensed. 3.7kApache-2.0Python
Hive (Aden)Production-grade multi-agent orchestration framework with 10K+ stars. Apache 2.0 licensed. 10kApache-2.0Python
Agent Squad (AWS Labs)Flexible multi-agent orchestration framework with intelligent intent classification and context management. Supports Python and TypeScript with pre-built agents for Bedrock, Lex, and custom integrations. Apache 2.0 licensed. 7.6kApache-2.0Python
DeepResearchAgentHierarchical multi-agent system for deep research tasks with automated task decomposition and execution across complex domains. 3.4kMITPython
Composio Agent OrchestratorAgentic orchestrator for parallel coding agents. Plans tasks, spawns agents, and autonomously handles CI fixes, merge conflicts, and code reviews. MIT licensed. 7.4kMITTypeScript
Open Multi-AgentTypeScript-native multi-agent orchestration with multi-model teams and parallel execution. Automatically converts goals to task DAGs. MIT licensed. 6.3kMITTypeScript
BeeAI Framework (IBM)Production-ready multi-agent framework in Python and TypeScript. Features workflow orchestration, ACP/MCP protocol support, and deep watsonx integration. Part of Linux Foundation AI & Data program. 3.3kApache-2.0Python
AI TownDeployable starter kit for building virtual towns where AI characters live, chat and socialize. Inspired by Stanford's Generative Agents research with persistent agent memory and social interactions. MIT licensed. 10kMITTypeScript
Conductor OSSEvent-driven agentic orchestration platform providing durable and resilient execution engine for applications and AI agents. Battle-tested at Netflix, Tesla, LinkedIn, and J.P. Morgan with 30K+ stars. Apache 2.0 licensed. 32kApache-2.0Java
A2A ProtocolAgent2Agent (A2A) open protocol enabling communication and interoperability between opaque agentic applications. Donated to Linux Foundation by Google with 50+ technology partners. Apache 2.0 licensed. 24kApache-2.0Shell
777genius/agent-teams-aiMulti-agent orchestration runtime with Kanban-style team management, message review loops, and provider integrations for reusable agentic teams. 1.2kAGPL-3.0TypeScript
Panniantong/Agent-ReachReusable search and web-ingestion layer for AI agents spanning Reddit, X/Twitter, YouTube, GitHub, Bilibili and more through one CLI. 24kMITPython
xerrors/YuxiSelf-hosted multi-tenant agent harness combining retrieval, knowledge graph grounding, and workflow orchestration for production teams. 5.5kMITPython
Sim StudioOpen-source AI workspace for building, deploying, and orchestrating AI agents. Visual canvas with 1000+ integrations, multi-framework support (Agno, OpenAI, LangChain, Google ADK), and self-hosted or cloud deployment. Apache 2.0 licensed. 29kApache-2.0TypeScript2FastLabs Agent SquadFlexible, lightweight open-source framework for orchestrating multiple AI agents to handle complex conversations with parallel execution capabilities. Apache 2.0 licensed. 7.6kApache-2.0Python
SIASelf-improving framework that orchestrates meta, target, and feedback agents to autonomously optimize the performance of AI models and agents on benchmark tasks. MIT licensed. 1.6kMITPython
Agent FileOpen file format (.af) for serializing stateful AI agents with persistent memory and behavior. Share, checkpoint, and version control agents across compatible frameworks. Apache 2.0 licensed. 1.2kApache-2.0TypeScript
Agent GatewayNext-generation proxy and routing layer for AI agents and MCP servers, with Kubernetes-native transport, protocol interoperability, and service-mesh-style isolation for reliable agent infrastructure. Apache 2.0 licensed. 3.1kApache-2.0RustAgent Governance ToolkitPolicy, safety, and execution controls for autonomous AI agents, including governance guardrails, sandboxing, and reliability checks. Apache 2.0 licensed. 4kMITPython
CodegraphLocal pre-indexed code knowledge graph for coding agents to reduce token usage and redundant tool calls across Claude, Codex, and other agents. 41kMITTypeScript
GortexLocal-first code knowledge graph and code intelligence engine built in Golang with multi-repository support and real-time graph actualisation. Built for AI coding agents aiming to expose only the needed information, reducing token usage by up to 50x. Works natively with Claude, Codex, Hermes, and other agents. 76Apache-2.0Go
GraphifyAI coding assistant skill that indexes codebases, databases, and documents into a queryable knowledge graph for coding agents. MIT licensed. 69kMITPython
HeadroomContext compression proxy for tool outputs, logs, and RAG chunks, reducing token pressure while preserving intent for AI agents. 12kApache-2.0Python
llmtrimSelf-hosted Rust proxy, MCP server, CLI, and library that compresses LLM prompts, tool outputs, and replies to cut token usage, quality-gated so it never raises your bill. MPL-2.0 licensed. 69AGPL-3.0Rust
MemPalaceHigh-performance, benchmarked AI memory system for persistent recall and retrieval in long-horizon autonomous workflows. 54kMITPython
SupermemoryMemory engine and API designed for long-lived AI agents to store, retrieve, and reuse long-horizon context with low latency. 25kMITTypeScript
HEXStrike AIMCP-powered coding-focused cybersecurity agent framework for automated pentesting and bug-hunting workflows. 9.2kMITPython
VulnClawAutonomous penetration testing agent utilizing Model Context Protocol (MCP) toolchains, blackboard state space search, and structured reasoning. 1.4kMITPython
OpenShell (NVIDIA)Safe and private runtime for autonomous AI agents with policy-driven execution boundaries and CLI integration. 6.8kApache-2.0RustCodeGraphPre-indexed local code knowledge graph for Claude Code, Codex, Gemini, and other coding agents to reduce context churn and token spend in developer workflows. 41kMITTypeScript
ECCPerformance-oriented agent harness for coding agents with skills, memory, and security-aware orchestration across Claude Code, Codex, and more. 207kMITJavaScript
oh-my-piTerminal coding agent with hash-anchored edits, subagents, LSP, browser integrations, and terminal-native workflows. 11kMITTypeScript
Pi (earendil-works)Modular agent toolkit with terminal-first CLI, unified model/provider layer, and runtime integrations for coding workflows and TUI/web UIs. 61kMITTypeScriptPi (badlogic)Terminal coding agent with hash-anchored edits, LSP integration, subagents, MCP support, and package ecosystem. 60kMITTypeScript
Mistral-Vibe (Mistral)Minimal CLI coding agent by Mistral. Lightweight, fast, and designed for local development workflows. 4.4kApache-2.0Python
Nanocoder (Nano-Collective)Beautiful local-first coding agent running in your terminal. Built for privacy and control with support for multiple AI providers via OpenRouter. 2k–TypeScript
Gemini CLI (Google)Open-source AI agent that brings Gemini's power directly into your terminal. Supports code generation, shell execution, and file editing with full Apache 2.0 licensing. 105kApache-2.0TypeScript
ArchonWorkflow engine for deterministic AI coding agents. Define development processes as YAML workflows (planning → implementation → validation → review → PR) with isolated Git worktrees for parallel execution. MIT licensed. 22kMITTypeScript
mini-SWE-agentLightweight coding agent for repository and issue-fixing workflows, designed for simple agentic software engineering experiments. 4.9kMITPythonTrae AgentSoftware-engineering agent from ByteDance for autonomous coding tasks and repository-level development workflows. 12kMITPython
Kilo CodeOpen-source agentic coding assistant with IDE workflows, tool use, and support for local or OpenAI-compatible models. 20kMITTypeScriptOpen SWEAsynchronous coding agent from the LangChain ecosystem for background software engineering tasks. 9.9kMITPythonLetta CodeMemory-first coding harness designed for long-lived agents that learn from experience. Persistent agents with portable memory across models (Claude, GPT, Gemini, GLM, Kimi). CLI and desktop app for macOS, Windows, and Linux. Apache 2.0 licensed. 2.7kApache-2.0TypeScript
gptmeYour agent in your terminal, equipped with local tools: writes code, uses the terminal, browses the web. Make your own persistent autonomous agent on top. MIT licensed. 4.3kMITPython
SuperpowersComposable skills framework and software development methodology for coding agents, structuring processes like planning, test-driven development, and code review. 226kMITShell
Agent SkillsProduction-grade engineering skills and quality gates for AI coding agents, packaging developer workflows like spec refinement, planning, and testing. 56kMITShell
HarnessTeam-architecture factory for Claude Code that designs domain-specific agent teams, defines specialized agents, and generates their skills. Apache 2.0 licensed. 6.9kApache-2.0HTML
jcodePerformance-oriented, memory-efficient coding agent harness built for multi-session workflows and infinite customizability. MIT licensed. 7.5kMITRust
DeepSeek-ReasonixDeepSeek-native AI coding agent for the terminal, designed around prefix-cache stability. MIT licensed. 24kMITGo
gstackMulti-specialist agent skills framework for Claude Code and coding agents that structures development workflows into planning, design, QA, and release phases. MIT licensed. 114kMITTypeScript
taste-skillAnti-slop frontend design skills for coding agents that improve layout, typography, and design system alignment during generation. MIT licensed. 50kMITJavaScript
OutlinesStructured outputs for LLMs. Guarantees valid JSON, regex-compliant text, and Pydantic model outputs during generation. Trusted by NVIDIA, Cohere, Hugging Face, and vLLM. Apache 2.0 licensed. 14kApache-2.0Python
PromptifyTask-based NLP engine with Pydantic structured outputs, built-in evaluation, and LiteLLM as the universal LLM backend. Think "scikit-learn for LLM-powered NLP". Apache 2.0 licensed. 4.6kApache-2.0Python
LangGPTPioneering framework for structured and meta-prompt design. Battle-tested by thousands of users worldwide with 10,000+ stars. The most popular prompt engineering paradigm for creating reusable, maintainable prompt templates. Apache 2.0 licensed. 12kApache-2.0Jupyter Notebook
Prompt OptimizerAI prompt optimization tool with multi-round iterative improvements, dual-mode optimization for system and user prompts, and multi-model support. Available as web app, desktop app, Chrome extension, and Docker deployment. AGPL-3.0 licensed. 30k–TypeScript
GuidanceEfficient programming paradigm for steering language models. Control output structure with loops, conditionals, and regex constraints inline. Reduces latency and cost vs conventional prompting. MIT licensed. 21kMITJupyter NotebookXGrammarFast, flexible and portable structured generation engine. Default backend for vLLM, SGLang, TensorRT-LLM, and MLC-LLM with flexible grammar support and zero-overhead mask generation. Apache 2.0 licensed. 1.7kApache-2.0C++
LM Format EnforcerEnforce output format (JSON Schema, Regex, etc) of language models by filtering allowed tokens at each generation step. Compatible with Hugging Face, llama-cpp-python, and vLLM. MIT licensed. 2kMITPython
ComposioTool integration layer for AI agents with 1000+ toolkits, authentication management, and sandboxed workbench. Powers tool use across major frameworks. 29kMITTypeScript
gpt-researcherAutonomous agent that conducts deep online research on any topic. Generates comprehensive reports with citations by orchestrating web searches, content scraping, and synthesis. Apache 2.0 licensed. 28kApache-2.0Python
last30days-skillAI agent that researches and synthesizes topic trends from Reddit, X, YouTube, Hacker News, and prediction markets. MIT licensed. 46kMITPython
pm-skillsMarketplace of product management plugin skills and workflows for Claude Code, Codex, and Claude Cowork. MIT licensed. 21kMIT
claude-code-best-practiceCustom subagents, skills, environment configurations, and best practices for the Claude Code developer agent. 60kMITHTML
PPT MasterAI-driven multi-role agent skill for converting documents into editable PowerPoint presentations via SVG and DrawingML workflows. MIT licensed. 34kMITPython
PraisonAI24/7 AI employee team for automating complex challenges. Low-code multi-agent framework with handoffs, guardrails, memory, RAG, and 100+ LLM providers. 8kMITPython
Agent-S (Simular AI)Open agentic framework that uses computers like a human. SOTA on OSWorld benchmark (72.6%) for GUI automation and computer control. 12kApache-2.0Python
MobileAgent (Alibaba/X-PLUG)Powerful GUI agent family for autonomous mobile device control. Multimodal agent framework designed to operate smartphone apps through visual UI perception and reasoning. MIT licensed. 8.8kMITPythonUI-TARS Desktop (ByteDance)Open-source multimodal AI agent stack with native GUI agent capabilities. Desktop application bringing GUI agent and vision power to your computer, browser, and terminal. Apache 2.0 licensed. 36kApache-2.0TypeScript
Browser UseMakes websites accessible for AI agents. Enables autonomous web automation, data extraction, and task completion with natural language instructions. MIT licensed. 97kMITPythonHeadroomToken and output compression middleware for AI agents and RAG workflows, reducing context footprint while preserving semantics (claims up to 60-95% fewer tokens). Apache 2.0 licensed. 12kApache-2.0Python
Steel BrowserOpen-source browser API for AI agents and apps. Batteries-included browser sandbox for web automation without infrastructure worries. Apache 2.0 licensed. 7.1kApache-2.0TypeScript
CuaOpen-source sandboxes, SDKs, and benchmarks for computer-use agents to control desktop environments. MIT licensed. 18kMITHTMLWebwrightA terminal-style web agent framework that enables coding models to run as browser agents by writing and executing Playwright scripts. MIT licensed. 5.7kMITPython
TradingAgentsMulti-agent framework for financial trading. Simulates professional trading firm operations with 6+ specialized agent roles, backtesting, risk management, and portfolio optimization. Built with LangGraph, supports multiple LLM providers. 83kApache-2.0Python
ParlantConversational control layer for customer-facing AI agents. Enterprise-grade context engineering framework optimized for consistent, compliant, and on-brand B2C and sensitive B2B interactions. Apache 2.0 licensed. 18kApache-2.0Python
n8nSelf-hostable workflow automation platform with AI agent nodes, tool integrations, and production automation workflows. 191k–TypeScript
ActivepiecesOpen-source automation platform with AI agents, MCP integrations, and self-hosted workflow orchestration. 23k–TypeScript
JulepStateful agent workflow platform with memory, tools, branching, and long-running task execution. 6.6k–Jupyter Notebook
uAgents (Fetch.ai)Fast and lightweight framework for creating decentralized agents with ease. Agents automatically join the network by registering on the Almanac smart contract. Supports agent-to-agent communication out of the box. Apache 2.0 licensed. 1.6kApache-2.0Python
TracecatSelf-hostable security automation platform for building agentic workflows across alerts, cases, and operations. 3.6kAGPL-3.0Python
ToolJetSelf-hostable internal app builder with AI app and agent workflows for operations teams. 38kAGPL-3.0JavaScript
Letta (ex-MemGPT)Platform for building stateful agents with advanced memory that learn and self-improve over time. 23kApache-2.0Python
Mem0Universal memory layer for AI agents. Persistent, multi-session memory across models and environments. 58kApache-2.0Python
Rohitg (agentmemory)Open-source memory service for agents with benchmarked retrieval, structured entities, and API/SDK support for persistent personalized memory in tooling workflows. 22kApache-2.0TypeScript
ForgetfulMCP server for persistent AI agent memory with atomic notes, semantic linking, and SQLite or PostgreSQL storage. 273MITPython
HindsightState-of-the-art long-term memory for AI agents by Vectorize. Fully self-hosted, MIT-licensed, with integrations for LangChain, CrewAI, LlamaIndex, Vercel AI SDK, and more. 16kMITPythonSupermemoryMemory API and app for AI agents that provides fast, scalable, context-aware storage and retrieval across projects. MIT licensed. 25kMITTypeScript
RAG open-source AI projects
Retrieval systems, vector databases, embedding models, and related tooling for RAG pipelines.
NornicDBLow-latency graph and vector hybrid retrieval database in Go with Neo4j and Qdrant-compatible drivers. 763MITGo
FaissSimilarity search and clustering library for dense vectors with CPU and GPU implementations. 40kMITC++
VespaAI + Data platform with hybrid search (vector + keyword) and real-time indexing at scale. Battle-tested serving billions of queries daily. 6.9kApache-2.0Java
pgvectorscalePostgreSQL extension for scalable vector search with DiskANN algorithm. Complements pgvector with significantly faster search and higher recall at large scale. PostgreSQL licensed. 3kPostgreSQLRust
VectorChordScalable, fast, and disk-friendly vector search in Postgres. Successor to pgvecto.rs with production-grade performance and efficient storage. AGPL-3.0 licensed. 1.7k–Rust
QuickwitCloud-native search engine for observability. Open-source alternative to Datadog, Elasticsearch, Loki, and Tempo with native vector search support. 11kApache-2.0RustTantivyFull-text search engine library inspired by Apache Lucene and written in Rust. Powers Quickwit and other production search systems. 15kMITRust
Manticore SearchEasy to use open source fast database for search. Good alternative to Elasticsearch with SQL-like interface and vector search capabilities. 12kGPL-3.0C++
OpenSearchOpen-source distributed and RESTful search and analytics suite with native vector search. Enterprise-grade fork of Elasticsearch with k-NN plugin for semantic search at scale. 13kApache-2.0Java
MarqoMultimodal vector search for text, image, and structured data. End-to-end indexing and search with built-in embedding models. Apache 2.0 licensed. 5kApache-2.0Python
ValdHighly scalable distributed vector search engine. Cloud-native architecture with automatic indexing, horizontal scaling, and multiple ANN algorithm support. Apache 2.0 licensed. 1.7kApache-2.0Go
hnswlibHeader-only C++ library for fast approximate nearest neighbors with Python bindings. Supports CRUD operations and concurrent read/write - unique among ANN libraries. Powers many production vector databases. Apache 2.0 licensed. 5.2kApache-2.0C++
turbovecRust-native high-performance vector index with Python bindings optimized for fast ANN search and SIMD acceleration on modern CPUs. MIT licensed. 8.8kMITPython
sqlite-vecA vector search SQLite extension that runs anywhere. Extremely small, "fast enough" vector search written in pure C with no dependencies. Perfect for embedded and edge deployments. MIT/Apache-2.0 dual licensed. 7.7kApache-2.0CzvecLightweight, lightning-fast, in-process vector database from Alibaba. Built on Proxima (Alibaba's battle-tested vector search engine) for production-grade, low-latency similarity search. Apache 2.0 licensed. 9.8kApache-2.0C++
MeilisearchLightning-fast search engine API with AI-powered hybrid search. Features typo-tolerant full-text search combined with HNSW-based vector search for semantic retrieval. MIT licensed. 58k–Rust
TypesenseOpen source alternative to Algolia + Pinecone. Fast, typo-tolerant, in-memory fuzzy search engine with native vector search capabilities. GPL-3.0 licensed. 26kGPL-3.0C++
ElasticsearchDistributed search and analytics engine with native k-NN vector search, hybrid search, and dense vector indexing. Industry-standard for full-text search now with powerful semantic search capabilities. AGPL-3.0/Elastic-2.0 dual licensed. 77k–JavaApache SolrMature Lucene-based search platform with dense vector search, filtering, faceting, and hybrid retrieval patterns for production search-heavy RAG systems. 1.6kApache-2.0Java
RediSearchFull-text, secondary indexing, and vector similarity search for Redis deployments. Useful when retrieval needs low-latency Redis-native search. 6.2k–C
ParadeDBPostgres-native search and analytics engine for full-text, faceted, and hybrid retrieval without moving data out of PostgreSQL. 8.9kAGPL-3.0Rust
OramaLightweight search engine with full-text, vector, and hybrid search for browser, server, and edge applications. 10k–TypeScript
HelixDBGraph-vector database for retrieval systems that need relationship traversal alongside semantic search. 4.7kApache-2.0Rust
USearchFast single-file similarity search & clustering engine for vectors. Smaller and faster than FAISS with 20+ language bindings (C++, Python, JavaScript, Rust, Java, Go, etc.) and support for custom metrics. Apache 2.0 licensed. 4.1kApache-2.0C++Voyager (Spotify)Spotify's next-gen approximate nearest-neighbor search library for Python and Java. Up to 10x faster than Annoy with 4x less memory, designed for production use at billion-vector scale. Apache 2.0 licensed. 1.6kApache-2.0C++
Deep LakeAI Data Runtime for Agents with serverless PostgreSQL and multimodal datalake. Store and search vectors, images, text, videos, and more with LangChain/LlamaIndex integrations. Used by Intel, Bayer, Yale, and Oxford. Apache 2.0 licensed. 9.2kApache-2.0C++DiskANN (Microsoft)Graph-structured indices for scalable, fast, fresh and filtered approximate nearest neighbor search. Handles billion-vector datasets on a single node with SSD-based indexing. MIT licensed. 1.8kMITRustSPTAG (Microsoft)Distributed approximate nearest neighbor search library with high-quality vector index build and online serving toolkits. Powers Bing's vector search at trillion-vector scale. MIT licensed. 5kMITC++
nanoflannC++11 header-only library for fast nearest neighbor search with KD-trees. Zero dependencies, single-file integration, and 2-3x faster than FLANN with modern C++. BSD licensed. 2.6k–C++NMSLIBNon-Metric Space Library for efficient similarity search in generic non-metric spaces. Comprehensive toolkit for evaluating k-NN methods with support for exotic distance functions. Apache 2.0 licensed. 3.6kApache-2.0C++
VearchCloud-native distributed vector database for AI-native applications. Efficient similarity search of embedding vectors with horizontal scaling and real-time indexing. Apache 2.0 licensed. 2.3kApache-2.0Go
FastEmbed (Qdrant)Lightweight, fast Python library for embedding generation with ONNX Runtime. Supports text, sparse (SPLADE), and late-interaction (ColBERT) embeddings without GPU dependencies. Apache 2.0 licensed. 3kApache-2.0Python
EmbedAnythingMinimalist, highly performant multimodal embedding pipeline built in Rust. Memory-safe, modular, and production-ready for text, image, and audio embeddings with seamless vector DB integration. Apache 2.0 licensed. 1.3kApache-2.0RustText Embeddings Inference (Hugging Face)Blazing fast inference solution for text embedding models. High-performance extraction with token-based dynamic batching, Flash Attention, and support for FlagEmbedding, E5, GTE, and more. OpenAI-compatible API with Docker deployment. Apache 2.0 licensed. 4.8kApache-2.0Rust
EmbedChainUniversal memory layer for AI agents. Simple API to create RAG applications over any dataset with support for multiple vector stores, embedding models, and LLM providers. Apache 2.0 licensed. 58kApache-2.0Python
DoclingDocument processing toolkit for turning PDFs and other files into structured data for GenAI workflows. 61kMITPython
MinerUHigh-accuracy document parsing for LLM and RAG workflows. Converts PDFs, Word, PPTs, and images into structured Markdown/JSON with VLM+OCR dual engine. 66k–Python
MarkerFast, accurate PDF-to-markdown converter with table extraction, equation handling, and optional LLM enhancement for RAG pipelines. 36kGPL-3.0Python
LightRAGGraph-based RAG with dual-level retrieval system. Simple and fast with comprehensive knowledge discovery (EMNLP 2025). 36kMITPythonRAG-AnythingAll-in-One Multimodal RAG system for seamless processing of text, images, tables, and equations. Built on LightRAG. 21kMITPython
RAGLite (Superlinear)Python toolkit for RAG with DuckDB or PostgreSQL. Lightweight, efficient retrieval-augmented generation without heavy dependencies. MPL 2.0 licensed. 1.2kMPL-2.0Python
GPT-RAG (Azure)Enterprise RAG pattern for Azure OpenAI at scale. Secure, production-ready architecture using Azure Cognitive Search and Azure OpenAI LLMs for ChatGPT-style Q&A experiences. MIT licensed. 1.2kMITPython
LangChain4jJava library for integrating LLMs into Java applications. Implements RAG, tool calling (including MCP support), and agents with seamless integration into enterprise Java frameworks like Spring Boot. Apache 2.0 licensed. 12kApache-2.0JavaKernel Memory (Microsoft)Memory solution for users, teams, and applications. RAG pipelines with document ingestion, vector indexing, and natural language querying with citations. Supports multiple LLM providers and vector stores. MIT licensed. 2.2kMITC#
txtaiAll-in-one AI framework for semantic search, LLM orchestration and language model workflows. Embeddings database with customizable pipelines. 13kApache-2.0Python
Infinity (Embeddings Server)High-throughput, low-latency serving engine for text-embeddings, reranking, CLIP, and ColPali. OpenAI-compatible API. 2.8kMITPython
FlashRAGEfficient toolkit for RAG research with 40+ retrieval and reranking models, 20+ benchmark datasets, and optimized evaluation pipelines (WWW 2025 Resource). MIT licensed. 3.5kMITPython
DocsGPTPrivate AI platform for building intelligent agents and assistants with enterprise search. Features Agent Builder, deep research tools, multi-format document analysis, and multi-model support. MIT licensed. 18kMITPython
llmwareUnified framework for building enterprise RAG pipelines with small, specialized models. Optimized for AI PC and local deployment with 300+ models in catalog. Apache 2.0 licensed. 15kApache-2.0Python
AutoFlowGraph RAG-based conversational knowledge base tool built on TiDB Vector and LlamaIndex. Features Perplexity-style search with built-in website crawler. Apache 2.0 licensed. 2.8kApache-2.0TypeScript
KAG (OpenSPG)Knowledge Augmented Generation framework for logical reasoning and factual Q&A in professional domains. Builds on OpenSPG knowledge graph engine to overcome traditional RAG vector similarity limitations. Supports multi-hop reasoning with schema-constrained knowledge construction. Apache 2.0 licensed. 8.8kApache-2.0Python
ChonkieLightweight document chunking library for fast, efficient RAG pipelines. Memory-safe with multiple chunking strategies (semantic, token, recursive) and direct vector DB integration. MIT licensed. 4.1kMITPython
PageIndex (VectifyAI)Vectorless, reasoning-based RAG framework using document index structure. Achieves high accuracy without vector databases through intelligent context engineering and reasoning-based retrieval. MIT licensed. 33kMITPython
Kotaemon (Cinnamon)Open-source RAG-based tool for chatting with your documents. Hybrid RAG pipeline with full-text and vector retriever, re-ranking, and multi-modal capabilities. Clean Gradio-based UI with support for local and API-based LLMs. Apache 2.0 licensed. 25kApache-2.0Python
Reader (Jina AI)Convert any URL to LLM-friendly input with a simple prefix (r.jina.ai). Free service that extracts article content, removes clutter, and returns clean Markdown for RAG and agentic workflows. Apache 2.0 licensed. 11kApache-2.0TypeScriptUltraRAG (OpenBMB)First lightweight RAG framework based on Model Context Protocol (MCP) architecture. Low-code RAG pipeline builder with comprehensive evaluation system and DeepResearch capabilities. From Tsinghua THUNLP, NEUIR, OpenBMB, and AI9stars. Apache 2.0 licensed. 5.6kApache-2.0Python
Semantic RouterSuperfast AI decision-making layer for LLMs and agents. Uses semantic vector space to route requests using semantic meaning rather than waiting for slow LLM generations. Cuts routing time from seconds to milliseconds. MIT licensed. 3.6kMITPython
NeuriteFractal Graph-of-Thought mind-mapping for AI agents, web-links, notes, and code. Rhizomatic workspace blending chaos theory, graph theory, and fractal logic for creative thinking and RAG workflows. MIT licensed. 2.1kMITJavaScript
PathwayPython ETL framework for stream processing, real-time analytics, LLM pipelines, and RAG. Features 350+ connectors with always-in-sync data from SharePoint, Google Drive, S3, Kafka, PostgreSQL and more. BSL 1.1 license (becomes Apache 2.0 after 4 years). 63k–Python
Infinity (AI Database)AI-native database built for LLM applications with incredibly fast hybrid search of dense vector, sparse vector, tensor (multi-vector), and full-text. Powers RAGFlow's document engine. Apache 2.0 licensed. 4.5kApache-2.0C++
PrivateGPTPrivate document Q&A project for local and offline RAG workflows where data stays inside the user's environment. 57kApache-2.0Python
FastGPTKnowledge-base platform with RAG retrieval, document processing, visual AI workflows, and self-hosted deployment options. 28k–TypeScript
MaxKBSelf-hostable knowledge-base and agent platform for document ingestion, RAG pipelines, and enterprise assistant workflows. 21kGPL-3.0Python
DB-GPTSelf-hosted AI data assistant for private knowledge, database-aware conversations, and data-heavy RAG workflows. 19kMITPython
localGPTLocal document-chat project for private, on-device Q&A over files without sending data to external APIs. 22kMITPython
SurfSensePrivacy-focused NotebookLM-style workspace for teams to search, organize, and query knowledge with self-hosted RAG. 14kApache-2.0Python
Crawl4AILLM-friendly web crawler that turns websites into clean Markdown for RAG and agentic workflows. 68kApache-2.0Python
ScraplingAdaptive web scraping and crawling framework for robust structured extraction from pages to large-scale pipelines. 61kBSD-3-ClausePython
LightpandaMachine-first headless browser in Zig; rendering-free and ultra-lightweight for AI agent browsing. 31kAGPL-3.0Zig
Paperless-AIAutomated document analyzer for Paperless-ngx with RAG-powered semantic search across your document archive. 5.7kMITJavaScript
FirecrawlWeb Data API for AI - search, scrape, and interact with the web at scale. Clean markdown/JSON output with proxy rotation and JS-blocking handled automatically. 129kAGPL-3.0TypeScript
invisible-playwrightPlaywright wrapper for a stealth-patched Firefox 150 binary. Drop-in Browser object for AI agents that need to ingest web data from sites with anti-bot guardrails (reCAPTCHA, FingerprintPro, Cloudflare). Spoofing happens in C++ source, not via JS overrides. MIT (wrapper) + MPL-2.0 (patches). 1.2kMITPython
OpenDataLoader PDFAccessibility-aware PDF parser and conversion pipeline for AI-ready markdown and structured data workflows. 24kApache-2.0JavaMarkItDown (Microsoft)Python tool for converting files and office documents to Markdown. Supports PDF, PowerPoint, Word, Excel, images, audio, HTML, and more with OCR and transcription capabilities. MIT licensed. 144kMITPython
LiteParseLightweight document parsing toolkit for AI and RAG pipelines with PDF/OCR extraction and clean preprocessing defaults. 9.4kApache-2.0RustPaddleOCRLarge-scale OCR suite with detection, recognition, and layout analysis, used widely for document digitization and downstream RAG pipelines. 81kApache-2.0Python
aisuiteSimple, unified interface to multiple Generative AI providers. Use OpenAI, Anthropic, Google, and 10+ other providers with a standardized API similar to OpenAI's. Switch between models or providers with a single line of code. MIT licensed. 14kMITPython
Spring AIApplication framework for AI engineering in the Spring ecosystem. Unified API for LLMs, vector stores, and embedding models with seamless integration into Spring Boot applications. Supports RAG, tool calling, and structured outputs. Apache 2.0 licensed. 8.9kApache-2.0Java
RigRust library for building scalable, modular LLM-powered applications. Type-safe agent framework with unified LLM interface, built-in vector store integrations, and ergonomic abstractions for production AI systems. MIT licensed. 7.5kMITRust
AxTypeScript framework for building reliable AI applications. "Official" DSPy-inspired framework for TypeScript with type-safe LLM interactions, chain-of-thought reasoning, and structured output validation. Apache 2.0 licensed. 2.8kApache-2.0TypeScript
GenkitOpen-source framework for building full-stack AI-powered applications in JavaScript, Go, and Python. Built and used in production by Google's Firebase. Unified interface for integrating AI models from multiple providers with built-in RAG, tool calling, structured outputs, and developer tools. Apache 2.0 licensed. 6.1kApache-2.0TypeScript
ContextGemEffortless LLM extraction framework for documents. Powerful abstractions for building extraction workflows with automated dynamic prompts, data modeling, validation, and precise reference mapping. Apache 2.0 licensed. 1.8kApache-2.0Python
EinoThe ultimate LLM/AI application development framework in Go. Drawing from LangChain and Google ADK, designed to follow Go conventions with composable components for chains, agents, and workflows. Apache 2.0 licensed. 12kApache-2.0Go
Media open-source AI projects
Open-source models and applications for image, video, audio, and 3D generation and editing.
Stable Diffusion WebUI Forge - NeoActively maintained Forge-based Stable Diffusion web UI with the familiar extension-driven workflow. 1.3kAGPL-3.0PythonDiffusersPyTorch library for diffusion pipelines spanning image, video, and audio generation. 34kApache-2.0Python
SD.NextAll-in-one WebUI for AI generative image and video creation with multi-platform support, SDNQ quantization, and balanced CPU/GPU memory offload. 7.1kApache-2.0PythonQwen-Image (Alibaba)20B MMDiT image foundation model with state-of-the-art complex text rendering and precise image editing. Strong performance in Chinese text generation. Apache 2.0 licensed. 8kApache-2.0Python
stable-diffusion.cppProduction-ready C++ inference runtime for SD, Flux, and related diffusion models, optimized for CPU and GPU deployment. 6.2kMITC++
UpscaylFree and open-source AI image upscaler for Linux, macOS, and Windows. Uses Real-ESRGAN and Vulkan architecture to enhance images by reconstructing high-resolution details. Cross-platform desktop app with batch processing. AGPL-3.0 licensed. 46kAGPL-3.0TypeScript
SkyReels V2/V3 (Skywork)First open-source infinite-length film generative model using AutoRegressive Diffusion-Forcing. 7k–Python
Hyperframes (HeyGen)Open-source long video generation platform for cinematic and social video creation with timeline control, diffusion-based motion modules, and multimodal conditioning. 26kApache-2.0TypeScript
LTX-2 (Lightricks)Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model. 7.8k–Python
Open-Sora-Plan (PKU-YuanGroup)Reproduction of Sora with full open-source pipeline for text-to-video generation. MIT licensed. 12kMITPythonHelios (PKU-YuanGroup)Efficient long-video generation framework with 24GB VRAM support for up to 10,000 frames (5+ minutes) and 1280×768 resolution. Apache 2.0 licensed. 1.9kApache-2.0Python
Pixelle-Video (AIDC-AI)Text-to-video foundation model optimized for long coherent scenes and controllable generation workflows. Apache 2.0 licensed. 22kApache-2.0Python
ViralMintEnd-to-end short-video pipeline that scouts trends, transcribes competitors locally, and assembles captioned videos with AI scripts, voice, and stock footage. AGPL-3.0 licensed. 10AGPL-3.0Python
MoneyPrinterTurboAn end-to-end short-video generation pipeline that automates scripts, footage collection, voiceover, and subtitle synthesis. MIT licensed. 86kMITPython
OpenMontageAgentic video production platform that orchestrates scriptwriting, asset generation, voice synthesis, editing, and rendering. AGPL-3.0 licensed. 7.5kAGPL-3.0PythonViMaxMulti-agent video generation framework that orchestrates scriptwriting, storyboarding, character design, and temporal visual consistency for end-to-end video synthesis. MIT licensed. 9.7kMITPython
ACE-Step 1.5Local-first music generation model with broad hardware support across Mac, AMD, Intel, and CUDA devices. 11kMITPython
Magenta RealTime 2Open-weights live music model for streaming generation and real-time interaction. Apache 2.0 licensed. 1.4kApache-2.0Python
Stable Audio ToolsStability AI's open-source audio and music generative models. Latent diffusion model for generating audio conditioned on metadata and timing, providing faster inference times and creative control for sound effects and music production. MIT licensed. 3.8kMITPython
GPT-SoVITSFew-shot voice cloning with just 1 minute of voice data. Combines GPT and SoVITS architectures for high-quality TTS with cross-lingual support and emotional expression. MIT licensed. 58kMITPython
Real-Time Voice CloningClone a voice in 5 seconds to generate arbitrary speech in real-time. SV2TTS implementation with speaker encoder and vocoder for instant voice synthesis. MIT licensed. 60k–Python
Training open-source AI projects
Tools for model training, fine-tuning, synthetic data generation, and distributed training.
OumiFully open-source platform for the complete foundation model lifecycle - from data preparation and training to evaluation and deployment. Supports 100+ models with 200+ recipes for fine-tuning gpt-oss, Qwen3, DeepSeek-R1, and more. Apache 2.0 licensed. 9.3kApache-2.0Python
MarinOpen-source framework for foundation-model development with composable data, training, and evaluation pipelines for modern LLM research. 1.1kApache-2.0Python
ms-swiftUnified training framework for 600+ LLMs and 300+ MLLMs with CPT/SFT/DPO/GRPO (AAAI 2025). 14kApache-2.0Python
LLM FoundryDatabricks' training framework for composable LLM training with StreamingDataset and Composer. 4.4kApache-2.0Python
torchtunePyTorch-native library for post-training, fine-tuning, and experimentation with LLMs. 5.8kBSD-3-ClausePython
kohya_ssGradio-based GUI and CLI for training Stable Diffusion models (LoRA, Dreambooth, fine-tuning, SDXL). Provides accessible interface to Kohya's powerful training scripts. 12kApache-2.0PythonTRL (Transformers Reinforcement Learning)Official library for RLHF, SFT, DPO, ORPO. 19kApache-2.0Python
verlVolcano Engine Reinforcement Learning for LLMs with PPO, GRPO, REINFORCE++, DAPO (EuroSys 2025). 22kApache-2.0PythonNeMo-RLScalable toolkit for efficient model reinforcement with DTensor and Megatron backends. 1.7kApache-2.0Python
OpenRLHFEasy-to-use, scalable RLHF framework based on Ray. Supports PPO, GRPO, REINFORCE++, DAPO with vLLM integration and async training. Apache 2.0 licensed. 9.6kApache-2.0Python
LMFlowExtensible toolkit for finetuning and inference of large foundation models. Features RAFT alignment algorithm and comprehensive model support. Apache 2.0 licensed. 8.5kApache-2.0PythonXTunerA next-generation training engine built for ultra-large MoE models with efficient QLoRA and full-parameter fine-tuning. Apache 2.0 licensed. 5.1kApache-2.0Python
LudwigLow-code framework for building custom LLMs and deep neural networks. Declarative YAML configuration for training state-of-the-art models with PEFT/LoRA, 4-bit quantization, distributed training via Hugging Face Accelerate, and native Kubernetes support. Linux Foundation AI project. Apache 2.0 licensed. 12kApache-2.0PythonTorchTitan (PyTorch)PyTorch native platform for training generative AI models at scale. Showcases 4D parallelism (FSDP, tensor, pipeline, context) for LLM pretraining with 65%+ speedups over optimized baselines. BSD-3-Clause licensed. 5.4kBSD-3-ClausePython
VeOmni (ByteDance)Versatile framework for both single- and multi-modal pre-training and post-training. Model-centric distributed recipe zoo supporting text, vision, audio, and video models with unified training interface. Apache 2.0 licensed. 2kApache-2.0Python
H2O LLM StudioNo-code GUI framework for fine-tuning LLMs. Streamlined interface for SFT, reward modeling, and model deployment. Apache 2.0 licensed. 5kApache-2.0Python
TinyZeroMinimal reproduction of DeepSeek R1-Zero for countdown and multiplication tasks. Clean, accessible implementation for understanding RL-based reasoning training. Apache 2.0 licensed. 13kApache-2.0Python
PRIME-RLAgentic RL Training at Scale from Prime Intellect. Framework for large-scale reinforcement learning capable of scaling to 1000+ GPUs with fully asynchronous RL, FSDP2 training, and vLLM inference. Apache 2.0 licensed. 1.4kApache-2.0Python
slimeLLM post-training framework for RL Scaling from THUDM. Supports SFT and RL training with multi-turn compilation feedback, powering projects like TritonForge for automated GPU kernel generation. Apache 2.0 licensed. 6kApache-2.0Python
rLLMDemocratizing Reinforcement Learning for LLMs. Framework for training AI agents with RL featuring near-zero code changes, CLI-first workflow, and 50+ built-in benchmarks. Supports GRPO, REINFORCE, RLOO with verl and tinker backends. Apache 2.0 licensed. 5.6kApache-2.0Python
EasyR1Efficient, scalable, multi-modality RL training framework based on veRL. Extends veRL to support vision-language models with GRPO algorithm for efficient RL training. Apache 2.0 licensed. 5kApache-2.0PythonLeRobotMaking AI for robotics more accessible with end-to-end learning. State-of-the-art approaches for imitation learning and reinforcement learning with pretrained models, datasets, and simulated environments. Apache 2.0 licensed. 25kApache-2.0Python
AI-ToolkitUltimate training toolkit for finetuning diffusion models. Easy-to-use all-in-one training suite supporting FLUX.1, FLUX.2, Stable Diffusion, and video models with both GUI and CLI interfaces. Consumer-grade hardware friendly with comprehensive LoRA and full fine-tuning support. MIT licensed. 11kMITPython
OneTrainerOne-stop solution for all your Diffusion training needs. Supports FLUX, Stable Diffusion 1.5/2.x/3.x/SDXL, Würstchen, PixArt, Hunyuan Video and more. Features full fine-tuning, LoRA, embeddings, masked training, automatic backups, and TensorBoard integration. GPL-3.0 licensed. 3kAGPL-3.0Python
FluxGymDead simple FLUX LoRA training UI with LOW VRAM support (12GB/16GB/20GB). WebUI forked from AI-Toolkit with backend powered by Kohya Scripts. Combines simplicity of Gradio interface with flexibility of Kohya's powerful training scripts. GPL-3.0 licensed. 3.2kMITPython
MiniMindTrain a 64M-parameter LLM from scratch in just 2 hours for $3. Complete from-scratch implementation covering MoE, data cleaning, pretraining, SFT, LoRA, RLHF (DPO/PPO/GRPO), tool use, and model distillation. All core algorithms implemented in pure PyTorch without high-level abstractions. Educational framework for understanding LLM internals. Apache 2.0 licensed. 51kApache-2.0Python
FastChatOpen platform for training, serving, and evaluating large language model chatbots. Powers Chatbot Arena (lmarena.ai) serving 10M+ requests for 70+ LLMs. Includes training code for Vicuna, MT-Bench evaluation, and distributed multi-model serving with OpenAI-compatible APIs. Apache 2.0 licensed. 39kApache-2.0PythonPaddleNLPEasy-to-use and powerful LLM library built on Baidu's PaddlePaddle framework. Supports 100+ models with efficient training, compression, and high-performance inference on diverse hardware. Features RsLoRA+ algorithm, DeepSeek V3/R1 support with FP8/INT8 quantization, and unified checkpointing. Apache 2.0 licensed. 13kApache-2.0Python
DataTrove (Hugging Face)Platform-agnostic data processing pipelines for LLM training at scale. Handles filtering, deduplication, and tokenization on local machines or SLURM clusters. 3.1kApache-2.0Python
Megatron-LMDistributed training framework and reference codebase for large transformer models at scale. 17k–Python
Nanotron (Hugging Face)Minimalistic 3D-parallelism LLM pretraining with tensor, pipeline, and data parallelism. Designed for simplicity and speed. 2.7kApache-2.0Python
veScale (ByteDance)Hyperscale PyTorch distributed training with flexible FSDP implementation for LLMs and RL training at scale. 1kApache-2.0Python
RLinfScalable open-source RL infrastructure for post-training foundation models via reinforcement learning. Features M2Flow paradigm for embodied AI and agentic workflows with real-world robotics integrations. Apache 2.0 licensed. 3.6kApache-2.0Python
LLM Compressor (vLLM)Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM. Supports GPTQ, AWQ, SmoothQuant, AutoRound, and FP8/INT8 quantization with seamless Hugging Face integration. 3.3kApache-2.0PythonNVIDIA Model OptimizerUnified library of SOTA model optimization techniques including quantization, pruning, distillation, and speculative decoding. Compresses deep learning models for deployment with TensorRT-LLM, TensorRT, and vLLM to optimize inference speed across NVIDIA hardware. 2.9kApache-2.0PythonMLOps open-source AI projects
Tooling for tracking, deploying, monitoring, and operating AI systems in production.
ClearMLOpen-source platform for experiment tracking, orchestration, data management, and model serving. 6.7kApache-2.0Python
AimSelf-hosted ML experiment tracker designed to handle 10,000s of training runs with performant UI and SDK for programmatic access. Apache 2.0 licensed. 6.1kApache-2.0Python
FeastOpen source feature store for ML. Manages offline/online feature storage with point-in-time correctness to prevent data leakage. Apache 2.0 licensed. 7.1kApache-2.0Python
OpenLineageOpen standard for lineage metadata collection designed to instrument jobs as they run. Defines a generic model of run, job, and dataset entities for consistent data lineage tracking. Apache 2.0 licensed. 2.5kApache-2.0Java
MarquezLF AI & Data Foundation Graduated project for metadata collection, aggregation, and visualization. Maintains provenance of how datasets are consumed and produced with global visibility into job runtime and dataset lifecycle management. Integrates with OpenLineage. Apache 2.0 licensed. 2.2kApache-2.0Java
CivitaiOpen-source AI model hub and community platform for sharing and discovering generative AI models, with focus on image generation models. Features model versioning, reviews, and integrated inference. Apache 2.0 licensed. 7.1kApache-2.0TypeScriptHugging Face HubOfficial Python client for the Hugging Face Hub. Download, upload, and manage 1M+ open-source ML models and datasets programmatically. The de facto standard for model sharing and distribution. Apache 2.0 licensed. 3.7kApache-2.0PythonModelScopeModel-as-a-Service platform bringing together 700+ state-of-the-art ML models from the AI community. Covers NLP, CV, Audio, Multi-modality, and AI for Science with streamlined model inference, fine-tuning and evaluation. Apache 2.0 licensed. 9kApache-2.0Python
OpenVINO Open Model ZooPre-trained deep learning models and demos optimized for Intel hardware. 200+ public pre-trained models for vision, speech, and NLP with benchmarking tools and accuracy metrics. Apache 2.0 licensed. 4.4kApache-2.0Python
ONNX Model ZooCollection of pre-trained, state-of-the-art models in the ONNX format. 80+ models spanning vision, NLP, and audio with validation data and reference implementations. Apache 2.0 licensed. 9.6kApache-2.0Jupyter NotebookTransformers.jsState-of-the-art Machine Learning for the web. Run Hugging Face Transformers directly in your browser with no server needed. Supports 1000+ models including BERT, GPT-2, T5, and more via ONNX Runtime Web. Apache 2.0 licensed. 16kApache-2.0JavaScript
DJL (Deep Java Library)Engine-agnostic deep learning framework for Java with built-in model zoo. Load and run PyTorch, TensorFlow, MXNet, and ONNX models with a unified API. Includes 80+ pre-trained models for CV and NLP. Apache 2.0 licensed. 4.8kApache-2.0JavaPaddleSegEasy-to-use image segmentation library with awesome pre-trained model zoo. Supports semantic segmentation, interactive segmentation, panoptic segmentation, image matting, and 3D segmentation with 200+ pre-trained models. Apache 2.0 licensed. 9.3kApache-2.0PythonTorchVision ModelsPyTorch's official computer vision library with 50+ pre-trained model architectures including ResNet, EfficientNet, Vision Transformers (ViT), ConvNeXt, and more. The de facto standard model zoo for PyTorch computer vision. BSD-3-Clause licensed. 18kBSD-3-ClausePythonTensorFlow Model GardenOfficial TensorFlow repository of state-of-the-art (SOTA) models and modeling solutions. Contains reference implementations for BERT, ResNet, Transformer, and many more with pre-trained weights and training scripts. Apache 2.0 licensed. 78k–Python
PINTO Model ZooRepository for storing models inter-converted between various frameworks. Supports TensorFlow, PyTorch, ONNX, OpenVINO, TFJS, TFTRT, TensorFlowLite (Float32/16/INT8), EdgeTPU, and CoreML. 4,100+ stars with extensive model conversion tools for edge deployment. MIT licensed. 4.4kMITPython
Cerebras Model ZooCollection of deep learning models and utilities optimized for Cerebras hardware. Includes reference implementations for Llama, Mixtral, DINOv2, and Llava with configuration files, data preprocessing tools, and checkpoint converters. 1,150+ stars. Apache 2.0 licensed. 1.2kApache-2.0Python
ZenMLPipeline and orchestration framework for taking ML and LLM systems from development to production. 5.4kApache-2.0Python
Seldon CoreMLOps and LLMOps framework for deploying, managing and scaling AI systems in Kubernetes. Standardized deployment across model types with autoscaling, multi-model serving, and A/B experiments. 4.8k–Go
MetaflowNetflix's ML platform for building and managing real-world AI systems. Powers thousands of projects at Netflix, Amazon, and DoorDash. Apache 2.0 licensed. 10kApache-2.0Python
FlyteKubernetes-native workflow orchestration platform for AI/ML pipelines. Dynamic, resilient orchestration with strong type safety and reproducibility. Used by Lyft, Spotify, and Gojek. Apache 2.0 licensed. 7.1kApache-2.0GoPrefectWorkflow orchestration framework for building resilient data and ML pipelines. Python-native with modern observability and 200+ integrations. Apache 2.0 licensed. 23kApache-2.0Python
DagsterCloud-native orchestration platform for developing and maintaining data assets including ML models. Declarative programming model with integrated lineage and observability. Apache 2.0 licensed. 16kApache-2.0Python
Kubeflow PipelinesMachine Learning Pipelines for Kubeflow. Platform for building and deploying portable, scalable ML workflows using Kubernetes and Argo. Apache 2.0 licensed. 4.2kApache-2.0Python
Argo WorkflowsCNCF graduated container-native workflow engine for orchestrating parallel jobs on Kubernetes. Powers Kubeflow Pipelines and widely used for ML/data processing at scale. Apache 2.0 licensed. 17kApache-2.0Go
MLRunOpen-source AI orchestration platform for quickly building and managing continuous ML and generative AI applications across their lifecycle. Automates data preparation, model tuning, and deployment. Apache 2.0 licensed. 1.7kApache-2.0Python
KestraEvent-driven orchestration and scheduling platform for mission-critical workflows. Infrastructure-as-Code approach with declarative YAML, Git version control integration, and hundreds of plugins for data pipelines and ML workflows. Apache 2.0 licensed. 27kApache-2.0Java
KitOpsCNCF open source DevOps tool for packaging, versioning, and securely sharing AI/ML models, datasets, code, and configuration. Packages everything into OCI artifacts stored in existing container registries. Apache 2.0 licensed. 1.3kApache-2.0Go
PolyaxonMLOps Tools For Managing & Orchestrating The Machine Learning LifeCycle. Reproducible and scalable machine learning workflows on Kubernetes with experiment tracking, model management, and pipeline orchestration. Apache 2.0 licensed. 3.7kApache-2.0Netflix MaestroNetflix's next-generation workflow orchestrator for data and ML pipelines at massive scale. Highly scalable and flexible scheduler designed to handle millions of workflows across thousands of nodes. Apache 2.0 licensed. 3.8kApache-2.0Java
HAMiHeterogeneous GPU Sharing on Kubernetes. CNCF sandbox project providing GPU virtualization, slicing, and scheduling for efficient AI workload management across heterogeneous accelerators (GPUs, NPUs, MLUs). Apache 2.0 licensed. 3.5kApache-2.0Go
NVIDIA KAI SchedulerKubernetes-native GPU scheduler for AI workloads at large scale. Originally developed by Run:ai, now open-sourced by NVIDIA. Optimizes GPU resource allocation with dynamic allocation and efficient queue management. Apache 2.0 licensed. 1.3kApache-2.0GoNVIDIA DeepOpsInfrastructure automation tools for building GPU clusters with Kubernetes and Slurm. Deploys multi-node GPU clusters with monitoring, logging, and storage for AI/HPC workloads. BSD-3-Clause licensed. 1.4kBSD-3-ClauseShell
SkyPilotRun, manage, and scale AI workloads on any AI infrastructure. Unified interface to access and manage compute across Kubernetes, Slurm, and 20+ cloud providers. Used by Shopify and research institutions for training and inference. Apache 2.0 licensed. 10kApache-2.0Python
VolcanoCloud-native batch scheduling system for compute-intensive workloads. CNCF incubating project with gang scheduling, job dependency management, and topology-aware scheduling for AI/ML and deep learning. Apache 2.0 licensed. 5.6kApache-2.0Go
FeaturetoolsOpen-source Python library for automated feature engineering. Transforms transactional and relational datasets into feature matrices for machine learning using Deep Feature Synthesis with reusable primitives. BSD-3-Clause licensed. 7.7kBSD-3-ClausePython
KedroToolbox for production-ready data science. Uses software engineering best practices to help you create data engineering and data science pipelines that are reproducible, maintainable, and modular. Apache 2.0 licensed. 11kApache-2.0Python
Feature-enginePython library with multiple transformers to engineer and select features for machine learning models. scikit-learn compatible with fit() and transform() methods for encoding, imputation, variable transformation, and feature selection. BSD-3-Clause licensed. 2.2kBSD-3-ClausePython
LiteLLMAI Gateway to call 100+ LLM APIs in OpenAI format with unified cost tracking, guardrails, load balancing, and logging. 49k–Python
OpenLITOpenTelemetry-native LLM observability platform with GPU monitoring, evaluations, prompt management, and guardrails. 2.5kApache-2.0TypeScript
OpenLLMetry (Traceloop)Open-source observability for GenAI/LLM applications based on OpenTelemetry with 25+ integration backends. 7.2kApache-2.0Python
AgentaOpen-source LLMOps platform combining prompt playground, prompt management, LLM evaluation, and observability. 4.2k–TypeScript
LatitudeOpen-source agent engineering platform with prompt management, evaluations, and optimization. Features prompt playground, LLM-as-judge evals, and GEPA prompt optimizer for production LLM features. LGPL-3.0 licensed. 4.1kMITTypeScript
HeliconeOpen-source LLM observability with request logging, caching, rate limiting, and cost analytics. 5.8kApache-2.0TypeScript
GiskardOpen-source evaluation and testing library for LLM agents. Red teaming, vulnerability scanning, RAG evaluation, and safety testing with modular architecture. Apache 2.0 licensed. 5.4kApache-2.0Python
Portkey GatewayBlazing fast AI Gateway to route 200+ LLMs with unified API. Integrated guardrails, load balancing, fallbacks, and cost tracking. MIT licensed. 12kMITTypeScript
Envoy AI GatewayManages unified access to generative AI services built on Envoy Gateway. Kubernetes-native AI gateway for routing, load balancing, and managing LLM traffic with enterprise-grade reliability. Apache 2.0 licensed. 1.7kApache-2.0Go
PezzoCloud-native LLMOps platform with prompt management, versioning, and observability. Features collaborative prompt editing, A/B testing, and cost analytics. Apache 2.0 licensed. 3.2kApache-2.0TypeScriptMicrosoft PromptFlowComprehensive suite for LLM-based AI app development from prototyping to production. Includes prompt engineering, evaluation, and deployment tools with VS Code integration. MIT licensed. 11kMITPython
ChainForgeVisual programming environment for battle-testing prompts and evaluating LLM outputs. Features node-based prompt chains, multi-model comparison, and hypothesis testing. MIT licensed. 3kMITTypeScript
Future AGIOpen-source self-hostable end-to-end agent engineering and optimization platform that unifies tracing, evals, simulations, datasets, gateway, and guardrails. Built for shipping self-improving AI agents with one feedback loop from prototype to production. Apache 2.0 licensed. 1.1kApache-2.0Python
PurpleLlama (Meta)Comprehensive set of tools to assess and improve LLM security. Includes Llama Guard safety classifiers, CyberSec Eval benchmarks, and Prompt Guard for prompt injection detection. BSD-3-Clause licensed. 4.2k–PythonGarak (NVIDIA)The LLM vulnerability scanner. Probes models for hallucinations, data leakage, prompt injection, misinformation, toxicity, and jailbreaks. Extensive plugin-based architecture with 100+ vulnerability probes. Apache 2.0 licensed. 8kApache-2.0Python
PromptfooOpen-source LLM evaluation and red teaming framework. Test prompts, agents, and RAGs with automated security vulnerability scanning, side-by-side model comparison, and CI/CD integration. Now part of OpenAI. MIT licensed. 22kMITTypeScript
Eval open-source AI projects
Benchmarks, evaluation frameworks, datasets, and supporting tools for model assessment.
LiveBenchContamination-free LLM benchmark with objective ground-truth scoring. ICLR 2025 spotlight paper featuring frequently-updated questions from recent sources. Tests math, coding, reasoning, language, instruction following, and data analysis. 1.2k–Python
MMLU-Pro / GPQAMore challenging MMLU-style benchmark suite for evaluating advanced language models with expert-level reasoning questions. 384Apache-2.0Python
OpenCompassEvaluation platform for benchmarking language and multimodal models across large benchmark suites. 7.1kApache-2.0Python
MLPerf InferenceIndustry-standard ML inference benchmarks with reference implementations for AI accelerators. 1.6kApache-2.0PythonMLPerf TrainingIndustry-standard ML training benchmarks from MLCommons. Reference implementations for training AI models at scale across image classification, object detection, NLP, and recommendation tasks. Apache 2.0 licensed. 1.8kApache-2.0PythonVLMEvalKitOpen-source evaluation toolkit for large multi-modality models (LMMs). Supports 220+ LMMs and 80+ benchmarks including MMMU, MathVista, and ChartQA. Powers the OpenVLM Leaderboard. Apache 2.0 licensed. 4.2kApache-2.0Python
Vectara Hallucination LeaderboardLeaderboard comparing LLM performance at producing hallucinations when summarizing short documents. Systematic evaluation of factual consistency across major models. Apache 2.0 licensed. 3.3kApache-2.0PythonSSWE-rebench (Nebius)Continuously updated benchmark with 21,000+ real-world SWE tasks for evaluating agentic LLMs. Decontaminated, mined from GitHub. ––
AgentBench (THUDM)Comprehensive benchmark to evaluate LLMs as agents across 8 diverse environments including household, web shopping, OS interaction, and database tasks. ICLR 2024. Apache 2.0 licensed. 3.5kApache-2.0Python
Inspect AIFramework for large language model evaluations from the UK AI Security Institute. 2.2kMITPython
LightevalEvaluation toolkit for LLMs across multiple backends with reusable tasks, metrics, and result tracking. 2.4kMITPythonOpenAI EvalsFramework for evaluating LLMs and LLM systems with an open-source registry of 100+ community-contributed benchmarks. MIT licensed. 19k–Python
LMMs-EvalUnified multimodal evaluation toolkit for text, image, video, and audio tasks with 100+ supported benchmarks. 4.2k–Python
BrowserGymGym environment for web task automation and agent evaluation. Includes MiniWoB, WebArena, WorkArena, and more. Apache 2.0 licensed. 1.2k–Python
TruLensEvaluation and tracking for LLM experiments and AI agents. Provides feedback functions for measuring quality, relevance, and groundedness with LangChain and LlamaIndex integrations. MIT licensed. 3.4kMITPythonOpenEvalsOpen-source evaluation library for LLM and agent applications. Built by LangChain with pre-built evaluators for common use cases including RAG, agents, and structured output validation. MIT licensed. 1.1kMITPython
AutoRAGRAG AutoML tool for automatically finding optimal RAG pipelines. Evaluates and optimizes retrieval-augmented generation with AutoML-style automation for your own data and use-case. Apache 2.0 licensed. 4.8kApache-2.0Python
E2B Code InterpreterPython & JS/TS SDK for running AI-generated code in secure isolated sandboxes. Essential infrastructure for evaluating code-generating LLMs with safe execution environments. Apache 2.0 licensed. 2.3kApache-2.0PythonSimpleEvals (OpenAI)Lightweight library for evaluating language models with transparent accuracy numbers. Reference implementations for MMLU, GPQA, MATH, HumanEval, MGSM, DROP, and SimpleQA benchmarks. MIT licensed. 4.5kMITPython
Safety open-source AI projects
Tools for alignment, interpretability, safety evaluation, and adversarial testing.
AgentOpsPython SDK for AI agent monitoring, LLM cost tracking, benchmarking, and evaluation. Integrates with CrewAI, Agno, OpenAI Agents SDK, LangChain, Autogen, AG2, and CamelAI. MIT licensed. 5.6kMITPython
BloomOpen-source agentic framework for automated behavioral evaluations of frontier AI models. Generates targeted evaluation suites to probe LLMs for specific behaviors (sycophancy, self-preservation, political bias, etc.) with quantitative elicitation rates. From Anthropic's safety research team. MIT licensed. 1.3kMITPython
interpret (Microsoft)Fit interpretable models and explain blackbox machine learning with state-of-the-art explainability techniques including Explainable Boosting Machines and SHAP-based explanations. 6.9kMITC++
nnsightLibrary for inspecting, tracing, and intervening on neural network internals at scale. 949MITPython
EasyEditEasy-to-use knowledge editing framework for LLMs. Enables precise modification of model knowledge and behavior to correct hallucinations or outdated information. ACL 2024. MIT licensed. 2.8kMITJupyter Notebook
AIX360Comprehensive AI explainability toolkit with interpretability algorithms for data and machine learning models. Includes TED, BRCG, and ProtoNN methods for diverse explanation needs. Apache 2.0 licensed. 1.8kApache-2.0Python
PyRIT (Microsoft)Python Risk Identification Tool for generative AI. Microsoft's open-source framework for automated red teaming with multi-modal attack support, crescendo strategies, and 100+ operations experience. MIT licensed. 3.9kMITPython
HereticOpen-source system for automatic censorship and robustness suppression removal in language-model outputs. 24kAGPL-3.0Python
Agentic SecurityAgentic LLM vulnerability scanner and AI red teaming kit with multi-step attack simulation and automated security probing. Apache 2.0 licensed. 1.9kApache-2.0PythonNeMo Guardrails (NVIDIA)Programmable guardrails toolkit for LLM-based conversational systems. Uses Colang DSL to define safety rules, dialog flows, and content boundaries. Integrates with LangChain, LangGraph, and LlamaIndex for production deployments. Apache 2.0 licensed. 6.4k–Python
Guardrails AIInput/output validation framework for building reliable AI applications. Detects and mitigates risks through composable validators for PII, toxicity, prompt injection, and structured output validation. Features Guardrails Hub with 50+ pre-built validators. Apache 2.0 licensed. 7kApache-2.0Python
DetoxifyTrained models and code to predict toxic comments on all 3 Jigsaw Toxic Comment Challenges. Built using PyTorch Lightning and Transformers for toxicity, severe toxicity, obscene, threat, insult, identity attack, and sexual explicit content detection. Apache 2.0 licensed. 1.3kApache-2.0Python
RedAmonAI-powered agentic red team framework that automates offensive security operations from reconnaissance to exploitation to post-exploitation with zero human intervention. Integrates multiple security tools for comprehensive penetration testing. MIT licensed. 2kMITPython
CAICybersecurity AI framework for semi- and fully-automating offensive and defensive security tasks. Purpose-built for cybersecurity use cases with agent-based architecture for vulnerability assessment and security operations. MIT licensed. 8.8k–Python
AI-Infra-Guard (Tencent)Full-stack AI Red Teaming platform securing AI ecosystems via OpenClaw Security Scan, Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation. Apache 2.0 licensed. 3.8kApache-2.0Python
PentestAgent (GH05TCREW)AI agent framework for black-box security testing, supporting bug bounty, red-team, and penetration testing workflows. MIT licensed. 2.6kMITPython
SuperagentProtects AI applications against prompt injections, data leaks, and harmful outputs. Embed safety directly into your app and prove compliance to your customers. MIT licensed. 6.6kMITTypeScript
Anthropic Sandbox RuntimePortable sandbox execution environment for AI agents with constrained filesystem and network boundaries, designed to reduce blast radius in tool-use and LLM autonomy tests. 4.4kApache-2.0TypeScript
SchemaBrainAn open-source schema intelligence and safety layer (like Cloudflare) between AI agents and databases. Exposes a read-only Model Context Protocol (MCP) interface with PII masking, structured query recovery, and tamper-evident SHA-256 audit logs. Apache 2.0 licensed. 3Apache-2.0Python
Domains open-source AI projects
NVIDIA ModulusOpen-source deep learning framework for physics-informed machine learning (Physics-ML). Build, train, and fine-tune models for AI4science and engineering applications using state-of-the-art SciML methods. Apache 2.0 licensed. 2.9kApache-2.0Python
BoltzOpen-source biomolecular interaction prediction models. Boltz-1 was the first fully open source model to approach AlphaFold3 accuracy; Boltz-2 adds binding affinity prediction for drug discovery. MIT licensed. 4kMITPython
PyMCModern, comprehensive probabilistic programming framework in Python. Bayesian modeling with advanced MCMC sampling, variational inference, and seamless integration with ArviZ for visualization. Apache 2.0 licensed. 9.6k–Python
MONAIMedical Open Network for AI. End-to-end framework for healthcare imaging with state-of-the-art, production-ready training workflows. Apache 2.0 licensed. 8.2kApache-2.0Python
Unity ML-AgentsToolkit for training intelligent agents in games and simulations using deep reinforcement learning. Enables NPC behavior control, automated testing, and game design evaluation. Apache 2.0 licensed. 19k–C#TianshouAn elegant PyTorch deep reinforcement learning library with clean API design and comprehensive algorithm implementations. Supports both single-agent and multi-agent RL with GPU acceleration. MIT licensed. 11kMITPython
RL Baselines3 ZooA training framework for Stable Baselines3 reinforcement learning agents with hyperparameter optimization, pre-trained agents, and extensive benchmark environments. MIT licensed. 2.8kMITPython
skrlModular reinforcement learning library implemented in PyTorch, JAX, and NVIDIA Warp with support for Gymnasium, NVIDIA Isaac Lab, MuJoCo Playground, and other environments. MIT licensed. 1.1kMITPython
OpenBBFinancial data platform for analysts, quants and AI agents. Open-source investment research infrastructure with extensive data integrations. AGPL-3.0 licensed. 69k–Python
FinGPTOpen-source financial large language models. Democratizing financial AI with data-centric training pipeline and multiple model releases for trading, analysis, and robo-advising. MIT licensed. 20kMITJupyter NotebookFinRLFinancial reinforcement learning framework for quantitative trading. Deep RL library for stock trading, portfolio allocation, and market execution with pre-built environments and benchmarks. MIT licensed. 15kMITJupyter NotebookQlibAI-oriented quantitative investment platform from Microsoft. Supports diverse ML modeling paradigms including supervised learning, market dynamics modeling, and RL. Now equipped with RD-Agent for automated R&D process. MIT licensed. 44kMITPythonFinRobotOpen-source AI agent platform for financial analysis using LLMs. Multi-agent system with specialized agents for trading, analysis, and research. Apache 2.0 licensed. 7.2kApache-2.0Jupyter Notebook
KronosFoundation model for financial candlesticks (K-lines) pre-trained on multi-dimensional market data across global exchanges. MIT licensed. 31kMITPython
CVATIndustry-leading data annotation platform for computer vision. Interactive video and image annotation tool used by tens of thousands of teams for machine learning at any scale. 16kMITPythonSAM 2Promptable image and video segmentation model with released checkpoints and training code. 19kApache-2.0Jupyter Notebook
Roboflow SupervisionReusable computer-vision utilities for detection, tracking, and segmentation pipelines in Python. 41kMITPython
Open3DModern library for 3D data processing with Python and C++ APIs. Core features include 3D data structures, processing algorithms, scene reconstruction, surface alignment, 3D visualization, and GPU acceleration. MIT licensed. 14k–C++
Point Cloud Library (PCL)Standalone, large-scale open project for 2D/3D image and point cloud processing. Comprehensive algorithms for filtering, feature estimation, surface reconstruction, registration, model fitting, and segmentation. BSD licensed. 11k–C++PyTorch3DFAIR's library of reusable components for deep learning with 3D data. Provides efficient 3D operators, differentiable rendering, and mesh processing tools integrated with PyTorch. BSD licensed. 9.9k–Python
RTAB-MapReal-Time Appearance-Based Mapping library for RGB-D, Stereo and LiDAR SLAM. Graph-based SLAM approach with incremental appearance-based loop closure detection for large-scale and long-term operation. BSD licensed. 3.8k–C++
MuJoCoGeneral-purpose physics simulator for robotics, biomechanics, and ML research. High-fidelity contact dynamics with native Python and C++ bindings. Apache 2.0 licensed. 14kApache-2.0C++
GluonTS (AWS Labs)Probabilistic time series modeling with deep learning. Powers Amazon SageMaker forecasting with PyTorch and MXNet backends. Apache 2.0 licensed. 5.2kApache-2.0Python
ExecuTorchPyTorch runtime and toolchain for deploying AI models on mobile, embedded, and edge devices. 4.7k–PythonApache TVMOpen Machine Learning Compiler Framework. Universal deployment to bring models into minimum deployable modules that can be embedded and run everywhere from datacenter to edge devices. Apache 2.0 licensed. 13kApache-2.0PythonNCNNHigh-performance neural network inference framework optimized for mobile platforms. No third-party dependencies, cross-platform, and runs faster than all known open-source frameworks on mobile CPU. Powers Tencent apps including QQ, WeChat, and Pitu. BSD-3-Clause licensed. 23k–C++
CARLAOpen-source simulator for autonomous driving research. High-fidelity simulation of urban environments with realistic physics, sensors, and traffic scenarios. Widely used for training and validating self-driving algorithms. MIT licensed. 14kMITC++
WebotsOpen-source multi-platform robot simulator providing a complete development environment for modeling, programming, and simulating robots, vehicles, and mechanical systems. Used in education, research, and industry. Apache 2.0 licensed. 4.4kApache-2.0C++Habitat-SimHigh-performance physics-enabled 3D simulator for embodied AI research. Supports 3D scans of indoor/outdoor spaces, CAD models, and configurable sensors. Powers Meta's embodied AI research. MIT licensed. 3.7kMITC++
Genesis WorldGeneral-purpose simulation platform for embodied AI and robotics research with flexible scenes, sensors, and agent workflows. Apache 2.0 licensed. 29kApache-2.0Python
UI open-source AI projects
OpenClawLocal-first personal AI assistant with multi-channel integrations and full agentic task execution. 377k–TypeScript
text-generation-webuiWeb UI for running local LLMs with multiple backends, extensions, and model formats. 47kAGPL-3.0Python
KhojSelf-hostable personal AI assistant for search, chat, automation, and workflows over local and web data. 35kAGPL-3.0Python
NewelleGNOME/Linux desktop virtual assistant with integrated file editor, global hotkeys, and profile manager. 1.4kGPL-3.0Python
NextChatLight and fast AI assistant supporting Web, iOS, macOS, Android, Linux, and Windows. One-click deploy with multi-model support. MIT licensed. 88kMITTypeScript
big-AGIAI suite for power users with multi-model "Beam" chats, AI personas, voice, text-to-image, code execution, and PDF import. MIT licensed. 7kMITTypeScript
MorphicAI-powered search engine with a generative UI. Supports multiple AI providers (OpenAI, Anthropic, Google, Ollama) and search providers (Tavily, SearXNG, Brave). Features smart search modes, widgets, and image/video search. Apache 2.0 licensed. 8.9kApache-2.0TypeScript
LeonYour open-source personal assistant. Built around tools, context, memory, and agentic execution. Self-hosted, privacy-focused, and extensible. MIT licensed. 17kMITTypeScript
WillowOpen source, local, and self-hosted Amazon Echo/Google Home competitive voice assistant alternative with hardware support. Apache-2.0 licensed. 3kApache-2.0CCoPawYour Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple chat apps with easily extensible capabilities. Apache-2.0 licensed. 17kApache-2.0Python
Smart2BrainPrivacy-focused Obsidian plugin for AI-powered second brain functionality. Chat with your notes using local or remote LLMs including Ollama and OpenAI. MIT licensed. 1.1kMITTypeScript
CasibaseOpen-source enterprise-level AI knowledge base and agent management platform. Supports multiple LLM providers, RAG, and team collaboration. Apache-2.0 licensed. 5.2kApache-2.0Go
BionicGPTOn-prem ChatGPT replacement for teams with assistants, RAG, access controls, auditing, and enterprise deployment features. 2.3k–Rust
LocalAIOpen-source AI engine running LLMs, vision, voice, image, and video models on any hardware. Self-hosted OpenAI-compatible API. MIT licensed. 47kMITGo
OnyxFull-featured AI platform with Chat, RAG, Agents, and Actions. 40+ document connectors and every LLM support. MIT licensed (Community Edition). 30k–Python
biniouSelf-hosted webUI for 30+ generative AI models. Generate multimedia content with AI on your own computer, even without dedicated GPU (8GB RAM minimum). Works offline once deployed. GPL-3.0 licensed. 1.1kGPL-3.0PythonSelf-hosted AI Starter Kit (n8n)Open-source Docker Compose template to quickly set up a local AI environment. Curated by n8n, combines self-hosted n8n with Ollama, Qdrant, and PostgreSQL for secure, self-hosted AI workflows. Apache 2.0 licensed. 15kApache-2.0
CoAINext-generation multi-tenant AI one-stop solution with built-in admin and billing system. Enterprise-grade unified LLM gateway supporting 200+ models and 35+ providers. Apache-2.0 licensed. 9.2kApache-2.0TypeScript
PlaneOpen-source Jira, Linear, Monday, and ClickUp alternative. AI-powered project management platform with intelligent task triage, sprint planning, and automated workflows. AGPL-3.0 licensed. 50kAGPL-3.0TypeScript
Cherry StudioAI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs. AGPL-3.0 licensed. 47kAGPL-3.0TypeScript
DeepChatA smart assistant that connects powerful AI to your personal world. Built-in MCP and ACP support, multiple search engines, privacy-focused with local data storage. Apache-2.0 licensed. 5.9kApache-2.0TypeScript
ChatALLConcurrently chat with multiple AI bots to discover the best answers. Desktop app for comparing ChatGPT, Claude, Gemini, and 20+ LLMs side-by-side. Apache 2.0 licensed. 16kApache-2.0JavaScript
ChatboxPowerful desktop AI client for ChatGPT, Claude, and other LLMs. Cross-platform with modern UI. GPLv3 licensed (Community Edition). 40kGPL-3.0TypeScript
MaidFree and open-source Android app for interfacing with llama.cpp models locally and remote APIs (Anthropic, DeepSeek, Mistral, Ollama, OpenAI). MIT licensed. 2.5kMITTypeScript
DiveOpen-source MCP Host Desktop Application with dual Tauri/Electron architecture. Seamlessly integrates with any LLMs supporting function calling. MIT licensed. 1.8kMITTypeScript
PocketPal AIOpen-source app that brings small language models directly to your phone. Run AI 100% privately on iOS and Android with no cloud required. MIT licensed. 7.2kMITTypeScript
Hermes DesktopDesktop companion application for installing, configuring, and chatting with Hermes Agent. MIT licensed. 12kMITTypeScript
Craft AgentsOpen-source desktop and web interface for agentic workflows, featuring built-in MCP support, customizable API connections, and session sharing. Apache-2.0 licensed. 6.7kApache-2.0TypeScript
LiveKit AgentsFramework for building realtime voice AI agents with WebRTC transport, STT-LLM-TTS pipelines, and production-grade orchestration. Used by Salesforce Agentforce and Tesla. Apache-2.0 licensed. 11kApache-2.0Python
Dev Tools open-source AI projects
RalphAutonomous AI development loop for Claude Code with intelligent exit detection. Automates iterative coding workflows with self-monitoring capabilities. MIT licensed. 9.3kMITShell
NimbalystDesktop app for running multiple Codex and Claude Code AI sessions in parallel Git worktrees. Test, compare approaches and manage AI-assisted development workflows in one unified interface. MIT licensed. 1.2kMITTypeScript
NezhaCode editor for the AI agents era. Run multiple Claude Code and Codex agents across projects on your machine with an intuitive interface. GPL-3.0 licensed. 1.4kGPL-3.0TypeScript
Aider DeskPlatform for AI-powered software engineers. Desktop application that enhances the aider terminal experience with a modern UI. Apache 2.0 licensed. 1.3kApache-2.0TypeScript
ZedHigh-performance, multiplayer code editor with built-in AI features. From the creators of Atom and Tree-sitter. Native AI agentic editing with support for any LLM provider. GPL licensed. 85k–Rust
Code ServerRun VS Code on any machine anywhere and access it in the browser. Self-hosted cloud IDE with full extension support. MIT licensed. 78kMITTypeScript
GitpodCloud development environment platform with automated prebuilds, ephemeral workspaces, and support for any IDE. Self-hostable with open-source core. AGPL-3.0 licensed. 14kAGPL-3.0TypeScript
OnlookOpen-source AI-first design and React editing environment for visually building and modifying frontend applications. 26kApache-2.0TypeScript
DaytonaSecure elastic infrastructure for running AI-generated code. Self-hosted alternative to GitHub Codespaces with support for multiple IDEs, prebuilds, and any cloud provider. AGPL-3.0 licensed. 73kAGPL-3.0TypeScript
AI WorkdeckOpen-source AI-native IDE workspace for legal and document-heavy workflows — "VS Code for lawyers." Self-hosted with MCP agent orchestration, OCR, due-diligence risk flagging, evidence-chain management, WPS WebOffice integration, smart clipboard. Supports air-gapped deployment with Ollama + local storage. AGPL-3.0 licensed. 46AGPL-3.0Java
ClineOpen-source IDE coding agent that can edit files, run commands, and use tools with user approval. 63kApache-2.0TypeScript
AiderTerminal-based AI pair programmer. Edit code in your local editor and aider implements the changes. Supports multiple LLMs, voice coding, and automatic Git commits. Top scores on SWE Bench. Apache 2.0 licensed. 46kApache-2.0Python
Kimi CLIKimi Code CLI agent from Moonshot AI. Terminal-based coding assistant with advanced context understanding and multi-file editing capabilities. Apache 2.0 licensed. 8.9kApache-2.0PythonQwen CodeOpen-source AI agent for the terminal, optimized for Qwen series models. Multi-protocol provider support including OpenAI, Anthropic, Gemini, Alibaba Cloud, OpenRouter. Features agentic workflow with Skills and SubAgents. Apache 2.0 licensed. 25kApache-2.0TypeScript
Open NotebookOpen-source implementation of Notebook LM with multi-modal content support (PDFs, videos, audio, web pages). Features multi-speaker podcast generation, 18+ AI provider integrations, and full-text + vector search. Self-hosted with complete data sovereignty. MIT licensed. 25kMITTypeScript
Deta SurfPersonal AI notebook for organizing files and webpages with AI-generated notes. Local-first data storage, open data formats, and open model choice including local models. Cross-platform desktop app for research and thinking workflows. Apache 2.0 licensed. 3.4kApache-2.0TypeScript
FilePilot AILocal-first file intelligence app, CLI, and MCP server for searching, summarizing, tagging, deduplicating, and safely organizing local files for AI agent workflows. MIT licensed. 9MITPython
QuartoOpen-source scientific and technical publishing system built on Pandoc. Create dynamic content with Python, R, Julia, and Observable. MIT licensed. 5.7k–JavaScript
DrawdataDraw datasets from within Python notebooks. Interactive data visualization tool for creating and editing datasets directly in Jupyter environments. MIT licensed. 1.7kMITJavaScript
llama.vimLocal LLM-powered code completion plugin for Vim/Neovim using llama.cpp. Fast, privacy-first, no API key needed. 2kMITVim Script
CodeCompanion.nvimAI-powered coding assistant for Neovim. Inline code generation, chat, actions, and tool use with support for multiple LLM providers. 6.6kApache-2.0Lua
ProxyAILeading open-source AI copilot for JetBrains IDEs. Connect to any model in any environment with auto-apply, image chat, file references, web search, and customizable personas. Apache 2.0 licensed. 1.9kApache-2.0Kotlin
avante.nvimNeovim plugin that brings Cursor-like AI IDE features to Vim. Edit code with natural language, generate code from context, and chat with AI about your codebase. Apache 2.0 licensed. 18kApache-2.0Lua
SerenaPowerful MCP toolkit for coding agents providing semantic retrieval and editing capabilities. Integrates language servers for IDE-level code understanding. MIT licensed. 25kMITPython
vim-aiAI-powered code assistant for Vim and Neovim. Generate code, edit text, and have interactive conversations with GPT models. Supports custom roles, vision capabilities, and any OpenAI-compatible API. MIT licensed. 1.2kMITPython
windsurf.vimFree, ultrafast Copilot alternative for Vim and Neovim. AI-powered code completion with low latency and large context window. MIT licensed. 5.1kMITVim Script
Minuet AINeovim plugin offering code completion as-you-type from popular LLMs including OpenAI, Gemini, Claude, Ollama, Llama.cpp, Codestral, and more. GPL-3.0 licensed. 1.2kGPL-3.0LuaPeekaboomacOS CLI & MCP server enabling AI agents to capture screenshots and automate UI interactions. Visual question answering through local or remote AI models. MIT licensed. 4.6kMITSwift
SkillsA collection of custom productivity and engineering skills for Claude Code and other AI coding agents. MIT licensed. 126kMITShell
planning-with-filesPersistent file-based planning skill for AI coding agents to survive context loss and restarts with a deterministic completion gate. MIT licensed. 25kMITPython
Stop SlopA skill file for Claude Code and other agents to remove AI writing patterns and tells from generated prose. MIT licensed. 12kMIT
Understand AnythingTurns subdirectories, codebases, or knowledge bases into an interactive, AI-navigable knowledge graph and dashboard. MIT licensed. 57kMITTypeScript
Knowledge Work Plugins (Anthropic)Collection of role-specific plugins and sub-agents for Claude Cowork and Claude Code. Apache 2.0 licensed. 20kApache-2.0Python
Cursor PluginsOfficial Cursor plugins for developer tools, workflows, and SaaS integrations. MIT licensed. 2k–TypeScriptClaude Plugins MarketplaceOfficial, Anthropic-managed directory of high-quality plugins for Claude Code. Apache 2.0 licensed. 31kApache-2.0Python
Agent Toolkit for AWSOfficial AWS-supported MCP servers, skills, and plugins to help AI coding agents build, deploy, and manage applications on AWS. Apache 2.0 licensed. 1.3kApache-2.0Python
CavemanSkill and setup script for Claude Code, Codex, Gemini, Cursor, and other coding agents that reduces token output by instructing agents to communicate in concise, caveman-style prose. MIT licensed. 83kMITJavaScript
Chrome DevTools MCPOfficial Model Context Protocol (MCP) server from Google that enables coding agents to control and inspect a live Chrome browser for automation, debugging, and web performance analysis. Apache 2.0 licensed. 45kApache-2.0TypeScriptCodex Plugin for Claude CodeOfficial integration plugin for Claude Code that enables developers to run Codex code reviews, delegate tasks, and manage background coding jobs. Apache 2.0 licensed. 23kApache-2.0JavaScript
Assistant UIReact/TypeScript library for building production-grade AI chat interfaces. Drop-in components for streaming messages, tool calls, and multi-modal inputs. 10kMITTypeScript
RulerCentral AI agent rule registry. Manages and distributes rules for AI coding agents across projects. MIT licensed. 2.7kMITTypeScript
PR-Agent (Qodo)AI-powered code review agent for GitHub, GitLab, Bitbucket, and Azure DevOps. Automated PR analysis, improvement suggestions, and multi-platform deployment via CLI, GitHub Actions, or webhooks. AGPL-3.0 licensed. 11kApache-2.0Python
LLM (Simon Willison)CLI tool and Python library for interacting with dozens of LLMs via remote APIs or locally. Extensible plugin ecosystem, SQLite logging. Apache 2.0 licensed. 12kApache-2.0Python
AIChatAll-in-one LLM CLI in Rust featuring Shell Assistant, Chat-REPL, RAG, AI Tools & Agents. Supports 20+ providers. MIT/Apache 2.0 licensed. 10kApache-2.0Rust
aicommitsCLI that writes your Git commit messages for you with AI. Never write a commit message again. Supports multiple providers including OpenAI, Groq, xAI, Ollama, and LM Studio. MIT licensed. 9kMITTypeScriptCodex CLIOpenAI's lightweight coding agent that runs in your terminal. Code generation, file editing, and command execution with approval. Apache 2.0 licensed. 89kApache-2.0Rust
RepomixPowerful tool that packs your entire repository into a single AI-friendly file. Perfect for feeding codebases to LLMs with smart filtering and token counting. MIT licensed. 26kMITTypeScript
GitIngestReplace 'hub' with 'ingest' in any GitHub URL to get a prompt-friendly extract of a codebase. Optimized for Python ecosystem and data science workflows. MIT licensed. 15kMITPython
InstructorPython library for extracting structured, validated data from LLMs using Pydantic models. Handles validation, retries, and error handling with 15+ provider support. MIT licensed. 13kMITPython
MirascopePython toolkit for building LLM applications with automatic versioning, tracing, and cost tracking. The "LLM Anti-Framework" for developers who want control. MIT licensed. 1.5kMITPython
Context7Up-to-date code documentation for LLMs and AI code editors. Fetches latest docs and code examples directly into LLM context via MCP. Eliminates hallucinated APIs. MIT licensed. 57kMITTypeScript
Claude SquadManage multiple AI terminal agents like Claude Code, Codex, OpenCode, and Amp. Terminal multiplexer for AI coding agents with session management and parallel execution. AGPL-3.0 licensed. 7.7kAGPL-3.0Go
HerdrTerminal agent multiplexer and workspace manager with mouse-native split-panes and automatic agent state detection. AGPL-3.0 licensed. 6.1k–Rust
agenttraceLocal-first TUI for observing AI coding agent sessions across Claude Code, Codex CLI, Gemini CLI, Aider, Cursor exports, OpenCode, and more. 54MITGo
agentsviewLocal-first session intelligence and cost analytics dashboard for AI coding agents, supporting Claude Code, Codex, and other tools. MIT licensed. 2.1kMITGo
Uni-CLISelf-repairing CLI catalog that exposes web, desktop, Electron, and bridge tools as deterministic commands for AI agents. 145Apache-2.0TypeScript
OpenChamber Mobile BridgeOpenCode/OpenChamber helper for exposing devcontainer-based coding sessions to private Tailscale mobile access through a host bridge. MIT licensed. 0MITJavaScript
Vercel AI SDKProvider-agnostic TypeScript toolkit for building AI-powered applications and agents. Unified API for OpenAI, Anthropic, Google, and 20+ providers with first-class streaming, tool-calling, and structured output support. Apache 2.0 licensed. 25k–TypeScript
GitHub Copilot SDKMulti-platform SDK for integrating GitHub Copilot Agent into apps and services. Production-tested agent runtime with planning, tool invocation, and context management. Build Copilot-style agents without writing your own orchestration. MIT licensed. 8.9kMITJava
Learn open-source AI projects
AI Engineering from Scratch (rohitg00)Comprehensive curriculum covering machine learning, deep learning, NLP, computer vision, and agents by implementing them from scratch. MIT licensed. 31kMITPython
Prompt Engineering Guide (DAIR-AI)Comprehensive guides, papers, lessons, and notebooks for prompt engineering, context engineering, RAG, and AI Agents. The definitive open-source resource for learning prompt engineering with 3M+ learners. MIT licensed. 75kMITMDX
Start Machine Learning (louisfb01)A complete guide to start and improve in machine learning and AI in 2026 without any background. Curated learning path with the latest news, state-of-the-art techniques, and comprehensive resources for beginners. MIT licensed. 5.3kMIT
ML For Beginners (Microsoft)12-week, 26-lesson, 52-quiz classic machine learning course for beginners. Comprehensive curriculum covering regression, classification, clustering, and NLP with practical projects. 86kMITJupyter Notebook
LLM Course (Maxime Labonne)End-to-end course for getting into Large Language Models with roadmaps and Colab notebooks. Covers pre-training, fine-tuning, RLHF, quantization, and prompt engineering. 80kApache-2.0AI For Beginners (Microsoft)12-week, 24-lesson curriculum on Artificial Intelligence. Covers symbolic AI, neural networks, computer vision, NLP, and reinforcement learning with hands-on labs. 48kMITJupyter NotebookGenerative AI for Beginners (Microsoft)21 lessons covering generative AI fundamentals, prompt engineering, RAG applications, fine-tuning, and LLM app deployment with practical exercises. 112kMITJupyter NotebookData Science for Beginners (Microsoft)10-week, 20-lesson curriculum on data science fundamentals. Covers data preparation, visualization, modeling, and deployment with practical projects. 36kMITJupyter Notebook
Learn PyTorch for Deep Learning (Zero to Mastery)Comprehensive PyTorch deep learning course with hundreds of exercises and real-world projects. 18kMITJupyter Notebook
The Incredible PyTorchCurated list of PyTorch tutorials, papers, projects, and communities for deep learning researchers. 13kMITDeep RL Class (Hugging Face)Free deep reinforcement learning course with hands-on exercises and trained agent publishing to the Hugging Face Hub. 4.9kApache-2.0MDX
Practical RL (Yandex Data School)Comprehensive reinforcement learning course covering RL fundamentals, deep RL, policy gradients, actor-critic methods, and practical applications in the wild. The Unlicense. 6.5kUnlicenseJupyter NotebookNLP Course (Yandex Data School)YSDA course in Natural Language Processing with 2025 materials covering text classification, language models, transformers, and modern NLP techniques. MIT licensed. 11kMITJupyter Notebook
Large Language Model Notebooks CoursePractical hands-on course about Large Language Models and their applications. Covers Chatbots, Code Generation, OpenAI API, Hugging Face, Vector databases, LangChain, Fine Tuning, PEFT, LoRA, QLoRA. MIT licensed. 1.8kMITJupyter Notebook
Transformers Tutorials (Niels Rogge)Comprehensive tutorials and demos using the Hugging Face Transformers library for NLP, vision, and multimodal tasks. 12kMITJupyter Notebook
Made With ML (Goku Mohandas)End-to-end course on building production-grade ML systems with MLOps fundamentals, from design to deployment and iteration. 48kMITJupyter Notebook
Awesome Machine LearningThe definitive curated list of machine learning frameworks, libraries and software organized by language. Covers Python, C++, Java, JavaScript, and more with comprehensive coverage of the ML ecosystem. CC0-1.0 licensed. 73k–Python
