Superopenai currently caches the results of LLM calls only in memory. Is there a way to save the cache on disk for subsequent runs of the program? I don't usually have multiple identical LLM calls within one run of the program, but duplication across multiple program runs is frequent.
Superopenai currently caches the results of LLM calls only in memory. Is there a way to save the cache on disk for subsequent runs of the program? I don't usually have multiple identical LLM calls within one run of the program, but duplication across multiple program runs is frequent.