Kong is internet hosting its API Summit 2024 occasion, and has made bulletins about a number of enhancements to its platforms. Key highlights are updates to Kong Konnect, its API administration platform, which additionally embrace new variations of Kong Insomnia, Kong Gateway, and Kong AI Gateway.
The most recent updates to Kong Konnect assist firms additional put together their API infrastructure to deal with AI use instances.
“There isn’t a AI with out APIs, and the most recent model of Kong Konnect delivers the important infrastructure for each. We intention to provide companies the instruments to handle and scale their API visitors securely, serving to drive innovation sooner than ever earlier than,” mentioned Augusto Marietti, CEO and co-founder of Kong Inc. “Kong Konnect offers a unified API platform for constructing, working and governing GenAI purposes.”
This replace consists of Konnect Service Catalog, which offers a single supply of fact for APIs and providers, which helps firms handle shadow APIs by permitting them to eliminate undiscovered or unused APIs. The Service Catalog additionally incorporates a Scorecard that assesses how compliance a service is with outlined standards.
Konnect Devoted Cloud Gateways are additionally now out there on Azure and new AWS areas. In accordance with the corporate, this added availability permits organizations to deploy and handle their APIs throughout a number of clouds and keep enterprise-grade SLAs.
Kong Insomnia 10, which is an API design and testing software, provides limitless assortment runs, AI Runner for creating GenAI apps, and Invite Management, which ensures that API property can solely be accessed by approved customers.
Different options on this launch of Kong Konnect embrace Serverless Gateways, a centralized and cloud-based repository for managing API configurations, and superior API and AI analytics.
The replace additionally introduces variations 3.8 of each Kong Gateway and Kong AI Gateway. Kong Gateway 3.8 provides Incremental Configuration Updates, which considerably reduces reminiscence and CPU utilization, and enhanced OpenTelemetry assist.
Kong AI Gateway 3.8 provides a number of semantic reasoning capabilities, together with Semantic Caching, which acknowledges when totally different prompts signify the identical query, leading to sooner response occasions. For example, “how lengthy does it take to prepare dinner pasta?” and “how lengthy does it take to prepare dinner spaghetti?” could be responded to with the identical cached response.
In accordance with the corporate, the power to answer with cached responses cuts down on processing occasions and reduces computational overhead.
One other new function on this replace is Semantic Immediate Guard, which permits AI Gateway to grasp the “essence” of a request and block inappropriate prompts. Usually, these prompts are sorted out by figuring out particular key phrases, and this function eliminates the necessity for that.
The three.8 launch additionally introduces Semantic Routing, which selects probably the most related LLM primarily based on a developer’s necessities.
“Kong AI Gateway 3.8 is a big step ahead in enabling organizations to totally harness the facility of AI,” mentioned Marco Palladino, co-founder and CTO of Kong. “By introducing Semantic Intelligence into our AI Gateway, we’re addressing probably the most urgent challenges confronted by enterprises leveraging GenAI right this moment — velocity, value and safety — whereas dramatically bettering the developer expertise.”
