Open Source Frameworks
Discover 104 open source Framework repositories from GitHub, each with an analysis of what it does, key features, use cases and alternatives. Framework projects here are most often combined with Python, LLM and AI Agents. Last updated October 4, 2026.
104 repositories · updated October 4, 2026
EasyEdit: Edit Knowledge and Steer Large Language Models
EasyEdit is a framework for changing specific knowledge or behavior in large language models and evaluating the effects. It brings together multiple editing and inference-time steering methods for researchers and developers comparing approaches or testing targeted edits.
cactus: Run AI Inference on Phones and Wearables
Cactus is a C++ inference engine for running language, vision, and speech models on mobile and edge devices. It combines quantization, device-focused kernels, and optional cloud handoff for applications that need local inference with a fallback for harder queries.
webiny-js: Build a Self-Hosted CMS on AWS
Webiny is a TypeScript CMS framework deployed in your own AWS account using serverless services. It suits teams that need customizable content APIs, tenant isolation, and infrastructure control, but requires AWS and development skills.
wake: Test and Analyze Solidity Smart Contracts
Wake is a Python-based framework for testing, fuzzing, and static analysis of Solidity smart contracts. It suits Solidity developers and security teams that want automated checks, custom detectors, and IDE support in one tool.
lagent: Build LLM-Powered Agents in Python
Lagent is a Python framework for composing LLM agents, tools, memory, and multi-agent workflows. It suits developers building tool-using or collaborative agent applications who want synchronous and asynchronous interfaces.
modular: Build and Deploy AI Models with MAX and Mojo
Modular combines the MAX framework for AI development and deployment with Mojo, a programming language and compiler. It suits developers building model-serving systems, accelerator code, or software using Mojo.
big_vision: Train and Evaluate Large-Scale Vision Models
Google Research’s JAX and Flax codebase for training and evaluating vision and image-text models on GPUs and Cloud TPUs. It suits researchers running scalable experiments, but project-specific code may not stay compatible with the current core.
SaaS-Boilerplate: Build SaaS apps with Next.js
A TypeScript starter for building full-stack SaaS applications with Next.js. It bundles authentication, team tenancy, permissions, database access, localization, and testing so developers can start from an integrated foundation.
TextMachina: Build Datasets for Machine-Generated Text Tasks
TextMachina is a Python framework for generating and exploring datasets for machine-generated text detection, attribution, and boundary tasks. It helps researchers and developers combine text sources, language models, and configurable generation pipelines while checking for dataset quality and bias.
quarkus: Build Cloud-Native Java Applications
Quarkus is a Java framework for building container-oriented applications, from microservices to monoliths. It supports both JVM and native deployments and combines imperative and reactive development styles.
sixpack: Run A/B Tests Across Multiple Languages
Sixpack is a self-hosted A/B-testing framework with an HTTP API that lets applications in different programming languages share experiment assignments and conversion data. It includes an optional dashboard for reviewing results.
Agent-S: Automate Desktop Tasks Through a GUI Agent
Agent-S is a Python framework that uses screenshots, mouse clicks, and keyboard input to carry out natural-language tasks in desktop applications. It suits research and automation workflows that need an agent to interact with a real computer interface.