Help Center>
ModelArts>
FAQs>
Service Deployment>
Service Deployment>
Real-Time Services>
Can a New-Version AI Application Still Use the Original API?
Updated on 2023-11-22 GMT+08:00
Can a New-Version AI Application Still Use the Original API?
ModelArts supports multiple model versions and flexible traffic policies to smoothly gray upgrade model versions. After a service is modified to deploy a new-version model or its model version is upgraded, the original service prediction API remains unchanged.
To adjust a model version, perform the operations described in this section.
Prerequisites
- A service has been deployed.
- A new-version AI application has been created.
Procedure
- Log in to the ModelArts management console. In the navigation pane on the left, choose Service Deployment > Real-Time Services. The Real-Time Services page is displayed.
- Locate the target service and click Modify in the Operation column. The Modify Service page is displayed.
- In the AI Application and Configuration area, click Add AI Application Version and Configuration to add a new version.
Figure 1 Add AI Application Version and Configuration
- Set the traffic proportion of the two versions. Service calling requests are allocated based on the proportion. For details about other settings, see Parameters. After the setting, click Next.
- Confirm the information and click Submit.
Parent topic: Real-Time Services
Real-Time Services FAQs
- What Do I Do If a Conflict Occurs in the Python Dependency Package of a Custom Prediction Script When I Deploy a Real-Time Service?
- How Do I Speed Up Real-Time Prediction?
- Can a New-Version AI Application Still Use the Original API?
- What Is the Format of a Real-Time Service API?
- How Do I Check Whether an Error Is Caused by a Model When a Real-Time Service Is Running But Prediction Failed?
- How Do I Fill in the Request Header and Request Body of an Inference Request When a Real-Time Service Is Running?
- Why Cannot I Access the Obtained Inference Request Address from the Initiator Client?
- What Do I Do If Deploying a Service Failed Due to Insufficient Quota?
- Why Did My Service Deployment Fail with Proper Deployment Timeout Configured?
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
The system is busy. Please try again later.
For any further questions, feel free to contact us through the chatbot.
Chatbotmore
