[Feature]: GPTCache implementation for GPT Assistants

[yashaswi-nayak]

Is your feature request related to a problem? Please describe.

We are creating multiple GPT Assistants to support our customer service operations. Currently, there is only support for caching results via ChatCompletion. We would like to see a GPT Assistant implementation of GPTCache.

GPT Assistants will mainly answer multiple queries that are very common like

• Greeting a user/customer
• Generic FAQs - questions that have been answered previously (stored in cache)
• Out-of-Scope Answers that are stored in the cache ( using RAG, we limit the answers to the ones stored in the cache)

Over the course of time, we build a decent cache storage of all answers for user queries. These answers can be pulled from the cache directly instead of requesting Open AI every time. This would save us a lot of tokens and latency time.

Describe the solution you'd like.

I would like to see a solution where we have

• GPT Assistant adapter class in Open AI adapter
• Functionality to select assistant via assistant_id
• Functionality to create and update cache for a specific assistant

Describe an alternate solution.

No response

Anything else? (Additional Context)

No response