Catch up on MuleSoft at Dreamforce. See how our latest innovations supercharge developers and AI agents.
Watch on Salesforce+Supercharge your AI initiatives with MuleSoft and NVIDIA NIM microservices.
Learn more
The MuleSoft AI Connectors
Ramp your AI capabilities quickly. Easily embed the latest AI technologies into your existing IT infrastructure with a comprehensive set of connectors for MCP, A2A, and more.
Simplify Autonomous Agent Development.
The MuleSoft AI Connectors streamline the design, build, and management of AI agents . Unify a wide range of cutting edge AI tooling such as Large Language Models (LLMs) and Vector Stores, which enable developers to create enterprise-grade AI solutions with ease.
Bridge AI Models and Enterprise Systems.
Seamlessly integrate diverse AI functionalities into existing enterprise applications with the MuleSoft AI connectors. Facilitate the creation of intelligent workflows that can leverage both Salesforce and non-Salesforce data and systems.
Enhance AI Agent Capabilities & Communication.
Empower AI agents with advanced capabilities such as the Model Context Protocol (MCP) and Agent-to-Agent (A2A). Enable agents to interact with various data sources and perform complex actions within any system, forming the foundation of your multi-agent, IT evolution.
The Inference and Vector Connectors
Seamlessly integrate AI models and vector databases into your applications. The MuleSoft Inference Connector allows you to connect to various large language models (LLMs) for tasks like text generation and summarization with natural language processing , while the MuleSoft Vector Connector provides access to a wide range of external vector stores . Together, they enable powerful AI capabilities, such as semantic search and Retrieval-Augmented Generation (RAG) , by bridging the gap between your Mule applications, AI models, and vector databases.
Unleash the full potential of your AI initiatives with a unified strategy, brought to you by the collaboration between MuleSoft and NVIDIA. Using NVIDIA NIM™ microservices with the MuleSoft Inference Connector, you can securely connect enterprise data to dozens of GPU-accelerated LLM models, including NVIDIA Nemotron models, a family of open models, datasets, and technologies that unlock developers to build the most efficient and accurate specialized agentic AI to run anywhere. This combines MuleSoft's integration with NVIDIA to bridge business data with generative AI.
The Model Context Protocol (MCP) Connector
This connector leverages the Model Context Protocol (MCP) to facilitate agent-to-system communication by enabling AI clients to interact with APIs, connectors, and apps within MuleSoft. This helps foster the seamless integration of LLM applications with your APIs, simplifying complex endpoints into more manageable tools for agentic clients.
The Agent-to-Agent (A2A) Connector
The Agent2Agent (A2A Connector) enables inter-agent communication for LLM agents and other applications by implementing the A2A protocol. This connector allows agents to interact and delegate specialized tasks to other agents, facilitating seamless collaboration. The A2A protocol supports capability discovery, interaction negotiation, and secure information exchange, ensuring efficient and secure communication between agents.
The Einstein Connector
The Einstein Connector allows developers to design, develop, and manage AI agents within Anypoint Platform using the Einstein Trust Layer. The Einstein Trust Layer ensures secure, reliable, and ethical use of AI by monitoring inputs and outputs, managing data privacy, and providing governance tools to control and track AI operations.
The Agentforce Connector
The MuleSoft Agentforce Connector allows developers to integrate Agentforce with external systems via Anypoint Platform. It simplifies connecting MuleSoft applications with Agentforce, enabling new experiences and streamlining manual processes.
Generally Available
The MuleSoft AI Chain Connector
The MuleSoft AI Chain Connector empowers developers to design, develop, and manage AI agents directly within Anypoint Platform. It simplifies AI integration by providing seamless interaction with Large Language Models (LLMs) and other AI technologies for tasks like natural language processing and text generation.
Open-source
The MuleSoft AI Chain Project
The MuleSoft AI Chain (MAC) Project is an open-source project empowering developers to integrate advanced AI capabilities into the MuleSoft ecosystem.
Coming soon
The Amazon Bedrock Connector
This connector enables support for Amazon Bedrock in Mule applications. Amazon Bedrock is a fully managed service offering top foundation models (FMs) from leading AI companies. It simplifies generative AI application development and is serverless, eliminating infrastructure management requirements and ensuring secure integration with AWS services.
Learn more about how MuleSoft is investing in AI Connectivity.
MuleSoft MCP Support
Accelerate development by transforming your APIs and applications into agent-ready assets. Power your AI agents with real-time, relevant business context unique to your organization, thanks to MCP-enabled integrations.
MuleSoft A2A Support
Enable secure collaboration and coordinated agent action across complex environments by leveraging the best protocols, tools, and governance capabilities. Lay the foundation for scalable, multi-agent systems with A2A.
Integrated Agentic Experiences
AI is rapidly evolving, and the current trend, known as the third wave of AI, focuses on agentic capabilities where AI agents autonomously execute tasks. At MuleSoft, we support this wave through integrating AI with enterprise systems via APIs, and our MuleSoft AI Connectors.