Is your feature request related to a problem? Please describe.
Since there is a time limit to run any worker, the only way to load a real dataset is to call the embedding in a batch. Limit of 200 requests per hour basically restricts the use case to a PoC.
Describe the solution you'd like
I'd rather limit the amount of requests to a shorter timeframe, like 10 requests per second. That would allow better control over enhancement progress and increase the speed of embedding process.
Describe alternatives you've considered
Perhaps a long running embedding API that will handle the request as fast as it can.
Is your feature request related to a problem? Please describe. Since there is a time limit to run any worker, the only way to load a real dataset is to call the embedding in a batch. Limit of 200 requests per hour basically restricts the use case to a PoC.
Describe the solution you'd like I'd rather limit the amount of requests to a shorter timeframe, like 10 requests per second. That would allow better control over enhancement progress and increase the speed of embedding process.
Describe alternatives you've considered Perhaps a long running embedding API that will handle the request as fast as it can.