Azure SQL supports Graph models via NODE and EDGE tables.
All details can be found here:
Graph processing with SQL Server and Azure SQL Database
The sample available in the Jupyter Notebook is the same available in the official documentation, re-created here just for your convenience.
For a more realistic sample, use the MillionSong dataset sample:
https://github.com/Azure-Samples/millionsongdataset-sql
Make sure also to take a look at this Case Study, that highlights how Azure SQL Graph support + ColumnStore can offer better performance that NoSQL databases:
BKW Energie builds digital view of energy grid, optimizes maintenance and production