The service is an expansion of a nearly four-year old partnership between Apple and IBM in which the two companies create corporate apps together for Apple devices like the iPhone and iPad. They’ve already created apps that help healthcare workers better track medical records and apps for workers in to field to more easily share information they gather with colleagues.

The partnership capitalizes on Apple’s consumer technology-know how and IBM’s expertise in enterprise technology. The idea is for IBM to sell corporate apps co-created with Apple’s design team for business customers to use on mobile devices, and for Apple to then sell those iPhones and iPads to businesses.

Apple is increasingly trying to make headway in selling its devices to corporations, an area in which it has previously played a supporting role as it focused on the consumer market. But as the overall consumer smartphone market declines, Apple sees business customers—long the domain of companies like Microsoft and Dell Technologies—as a new area of growth.

Apple and networking giant Cisco have a similar partnership in the enterprise arena in which Cisco (CSCO) sells networking services and work collaboration software like videoconferencing tools that have been modified to work better with Apple devices.

IBM (IBM) and Apple (AAPL) declined to comment about any revenue from their partnership, only that they have generated “several billion dollars in customer signings.” Mahmoud Naghshineh, IBM’s general manager, of partnerships and alliances, said that IBM has sold over 800 corporate apps to businesses as a result of its Apple partnership. IBM also said that Apple and IBM have built 100 apps tailored for specific industries like travel, retail, and transportation.

In expanding the partnership, Apple and IBM are now not just developing business apps for customers. The new artificial intelligence service is intended for corporate developers to build apps themselves.

The service, Watson Services for Core ML, links Apple’s Core ML tools for developers that it unveiled last year with IBM’s Watson data crunching service. Core ML helps coders build machine learning-powered apps that can more efficiently perform calculations on smartphones instead of processing those calculations in external data centers. It’s similar to other smartphone-based machine learning tools like Google’s TensorFlow Lite.

“With IBM our goal has been to help enterprises reimagine the way they work,” Apple vice president of OS product marketing Brian Croll said in a statement. “We believe the combination of Core ML and Watson Services will inspire the next generation of intelligent enterprise apps.”

Naghshineh, from IBM, pitched Watson Services for Core ML as a way for companies to build more compelling smartphone apps than they could previously. He gave the example of businesses that employ field technicians who inspect power lines or machinery.

For example, Naghshineh said companies could feed images of electrical equipment to Watson in order to train it to recognize the machinery. The technology would then be added as an app to field technicians’ iPhones or iPads so they could scan the electrical equipment they are inspecting and automatically detect any anomalies.

It would eliminate the need to send that data to IBM’s cloud computing data centers to process, thus speeding up the amount of time it takes to detect equipment bugs to near real-time.

Apple’s Core ML toolkit could already be used to connect with competing cloud-based machine learning services from Google (GOOG), Amazon (AMZN), and Microsoft (MSFT). In this case, what IBM did was create developer tools to more easily link the Core ML service with Watson.

Coca-Cola is testing Watson Services for Core ML to see if it helps its field technicians better inspect vending machines, Naghshineh said without providing any additional details. IBM did not say how much Coca-Cola is paying for the technology, if anything, only that Coca-Cola’s participation has helped IBM develop and improve the technology before making it more widely available.

For a limited time, the service will be free to developers to use. Eventually, developers will have to pay, but there is no date set.

IBM Watson general manager David Kenny explained that the new service is part of a broader suite of Watson artificial intelligence services for corporate developers who are building machine-learning apps.

While IBM has been selling Watson data crunching for several years, Kenny said that many of those deals have involved IBM’s consulting teams helping businesses use the company’s machine learning services. The new services represent a shift for IBM and its consulting teams in that the new Watson developer services are intended to be bought in an “accessible and bite size” way and sold in a “pay as you go” model without any related consultants, Kenny said.

Get Data Sheet, Fortune’s technology newsletter.

Still, Naghshineh explained that IBM consultants would still play a role in bigger corporate deals. While developers can use the new service to build the machine-learning apps, companies will still need to use IBM consultants for creating their overall data analytics and software strategies.

“You don’t need consultants there to show up and say here’s how to build an app,” Naghshineh said.