Ability to use Anthropic's Claude and/or other LLM models
Just like there are use cases to use GPT3.5 vs GPT4, there are also use cases where Claude-2 or Claude-Instant could provide an improved output.
Just like there are use cases to use GPT3.5 vs GPT4, there are also use cases where Claude-2 or Claude-Instant could provide an improved output.
Log in to comment and vote
Comments3
My AskAI
Sep 30, 2023
So we actually do also use Claude in some instances where we feel it works best - e.g. summarize!
mparkerchavez
Sep 30, 2023
May I ask why it's not an option for some of the other modes?
My AskAI
Sep 30, 2023
Sure, because we have done quite a bit of testing with it for shorter questions and found the answer quality to be quite a bit worse than the GPT models based on our benchmarking tests.