Hello, I have code that is polling the Retrieve Vector Store API looking for a status=”complete”.
After observing status=”complete”, I go ahead and call the responses API with file_search and the corresponding vector store id. However oftentimes the model will respond with something like “the file doesn’t exist, etc” indicating that the vector store was not actually available. If I wait a little bit and query the model again, then it will respond as expected with content from the vector store. This seems to indicate that there is a delay between the api reporting status=”complete” and the vector store actually being available.
The documentation says that a vector store status of “complete” indicates that the vector store is ready to be used, however that doesn’t always seem to be the case.