Open Source Alternatives to Refresh

An open-source framework for building, operating, and benchmarking Multimodal Language Model (MLM) agents across diverse environments.

An open-source, end-to-end AI agent engineering platform for simulating, evaluating, and optimizing LLM applications.

BenchFlow is a frontier environment lab that builds environments for AI agents to learn and be evaluated on real computer work.

Experimental open-source framework for building self-building autonomous agents with a novel function framework.

Together AI offers an AI-native cloud platform providing high-performance inference APIs, GPU clusters, and fine-tuning capabilities for open-source AI models.

Open-source AI research lab training decentralized large language models and building infrastructure for accessible AI.

DeepSeek is a Chinese AI company developing open-weight large language models and an AI assistant, aiming to achieve Artificial General Intelligence.

Open-source autonomous AI agent platform that automates complex digital workflows.

A managed, private, and production-grade environment for OpenClaw, offering access to over 500 AI models and essential tools.

Cua provides an open-source framework and cloud desktops for AI agents to control operating systems and desktop applications across Linux, Windows, macOS, and Android.

Company

Simulates 1,000 LLM-powered agents for diverse movie recommendation preferences.

An open-source framework for on-machine AI agents developed by Block's Open Source Program Office.

Open-source AI platform for securely deploying intelligent agents, private enterprise search, and complex document workflows on an organization's own data.

An open-source platform for creating, managing, and interacting with AI workflows and agents.

An open-source TypeScript/JavaScript framework for building AI agents, from the creators of Gatsby.

Developing advanced AI models with a focus on reducing hallucinations, including large language models, vision language models, and text-to-video models.

An open-source AI and machine learning platform for enterprises.

Persistent memory layer for AI agents and applications.

An open-source feature store for managing and serving data for AI/ML models in production.

Track experiments, validate models, and collaborate with confidence using Skore, the methodology layer for AI coding tools.

Parlant is an open-source conversational AI engine dedicated to solving the Human/AI alignment problem by enabling Large Language Models to reliably perform as desired and prevent hallucinations.

Open-source Python framework for building autonomous AI agents and agentic workflows, purpose-built for FinTech operations.

Cisco's internal incubation engine for emerging technologies like agentic AI and quantum networking.

Open-source infrastructure, developer tools, and research for agent runtimes.

A unified data analytics and AI platform built on the lakehouse architecture.

A non-profit AI research lab focused on open science and developing advanced multimodal and voice AI models.

Open-source tool to transform application data into accessible knowledge for AI agents.

Open-access AI cloud platform providing affordable, on-demand GPUs for training and serving AI models.

Nomic AI builds tools to make AI explainable and accessible, focusing on data curation and visualization.

Elementary Data provides a unified data control plane for the AI era, offering observability, quality, governance, and discovery for data and AI reliability.

An open-source Python library providing a unified interface for interacting with various LLM providers.

Run large language models locally on your machine.

Open-source library for building hyper-realistic voice agents and conversational AI applications.

An AI-native workspace designed to turn team interactions into actionable, proprietary data for agentic workflows.

Arcee provides U.S.-made, open source large language models and checkpoints for insight into raw model intelligence.

Building general-purpose AI for a future of material abundance.

An open-source red-teaming toolkit for testing the security of LLM-powered agents, auditing prompt injection, memory poisoning, and tool abuse.

Best free & open source AI personal assistant.

A C/C++ tensor library for efficient machine learning inference on consumer hardware.

Open-source, offline-deployable RAG and Agent application built on LangChain for local knowledge base Q&A.

A community-driven marketplace for discovering and deploying Model Context Protocol (MCP) servers and clients.

OpenWorker is an open-source desktop AI agent that runs locally, integrates with everyday tools, and delivers finished outcomes by breaking down tasks.

High-performance, open-source vector search engine built in Rust for production-grade AI and machine learning workloads.

Open-source Postgres platform for agent scale with copy-on-write branching.