Kong Open Source New AI Gateway to Help Developers Easily Build Multi-LLM Apps
Kong Open Source New AI Gateway to Help Developers Easily Build Multi-LLM Apps
Kong Inc. has introduced a suite of open-source plugins for Kong Gateway 3.6, transforming it into a comprehensive AI Gateway. This update allows organizations to seamlessly integrate and orchestrate multiple Large Language Models without modifying application code. By centralizing these capabilities, the platform enables developers to rapidly deploy AI features while maintaining strict security, visibility, and compliance controls over all AI traffic. The release addresses critical infrastructure needs for AI adoption, including unified credential management and advanced prompt engineering. Key features like no-code transformers and prompt guardrails allow teams to enforce governance and adapt AI behaviors dynamically. This architecture significantly reduces the operational complexity of managing diverse LLM integrations, ensuring that enterprises can leverage industry-leading models while protecting sensitive data and adhering to regulatory standards. This development is highly relevant to open data principles as it democratizes access to powerful AI tools by releasing the underlying code publicly. Open-sourcing these plugins fosters transparency and community-driven innovation, allowing organizations to audit, modify, and contribute to the security and functionality of their AI gateways. It empowers the developer community to build more robust, accountable, and interoperable AI systems without relying on closed-source solutions, thereby promoting trust and collaboration in the open source ecosystem.
Source: pressreleases.responsesource.comPublished on 2024-02-16
Related news
- Kong targets democratization of multi-LLM usage through open-source AI Gateway
- Kong's new open source AI Gateway makes building multi-LLM apps easier | TechCrunch
- This German nonprofit is building an open voice assistant that anyone can use | TechCrunch
- A beginners guide to fine tuning LLM using LoRA