worn fern
Hi Team,
I am experiencing delayed response from the Wizard 8x22b model for the prompt specified from last couple of weeks.
Prompt consists of a detailed structure for extracting relevant information from a raw text and convert the same into an asked JSON format.
But the API is giving response in 1min+ timeframe.
Can you share possible reasoning on the same to resolve such issues?
wet osprey
would you like to provide us your example request?
worn fern
We are curating a JSON structure from specified schema to fetch information from candidate's resume. The Candidate Resume is sometimes 5-10 pages long.
mystic violet
just use another model then?
worn fern
We tried, but amongst all "Wizard" performed well in extracting all the information from the raw text.