Content
@
0 reply
0 recast
0 reaction
ken
@kenergy.eth
Started using perplexity a lot more once I added the perplexity app into raycast. Both awesome recommendations found from farcaster that are improving my workflow.
1 reply
1 recast
2 reactions
Vladyslav Dalechyn
@dalechyn.eth
How’s your experience overall using it comparing to gpt4?
1 reply
0 recast
1 reaction
ken
@kenergy.eth
1. Token/s is much faster - sometimes it feels like I'm waiting forever in gpt4, especially in long responses 2. Source links are much better. Sometimes gpt4 links don't really seem relevant 3. The LLM response quality itself is comparable 4. 'related questions' are useful Overall better experience for research
1 reply
0 recast
0 reaction
ken
@kenergy.eth
Update - I asked perplexity a query about what turned out to be a bit of a complex Zapier automation, so it needed to research both how Zapier works and also break down the logic of implementing it using the provided tools. It struggled with the logic. GPT4 provided a much better breakdown about how to do it.
0 reply
0 recast
0 reaction