Hacker News new | ask | show | jobs
by ainadan 999 days ago
Looks great! Could FastStream be suitable choice for running ML inferences with improved memory management and congestion control compared to REST APIs?
1 comments
davorrunje 999 days ago
Actually, serving ML inferences was the initial motivation for building the library. We couldn’t find anything similar to FastAPI for Apache Kafka and then we decided to build one.
link