Azure Machine Learning - General Availability for November

Five features now available in GA enable you to search thousands of OSS LLMs, provide a feature set specification and let the system handle the rest, and lower. compute management overhead and setup costs. You can also now deploy complex compute graphs under batch endpoints, and use one platform to build, tune, evaluate, deploy, and test AI workflows.  

Model Catalog: It allows you to discover, customize and operationalize large foundation models without the hassle of managing infrastructure or software dependencies. 

Managed Feature Store: You can now experiment and ship models faster, increase reliability of your models and reduce your operational costs. 

Model Training with Serverless Compute: You can now focus on building ML models without needing to learn how to set up compute infra. 

Deploy pipelines and components under Batch Endpoints: You can now deploy multiple versions of the pipelines using multiple deployments under the same endpoint and switch between them without disturbing your downstream consumers 

Build high quality intelligent applications with Prompt flow: You can now quickly build a workflow that connects different data sources with LLM for building intelligent applications.