@paraga is there any chance we can get first party support for P0 on Cloudflare's AI Gateway? Recently moved from P0 to PPLX over this, but the response quality just is not the same @Cloudflare @dok2001
1
2
2,791
Let's do it. @dok2001 can you please connect me to the right people on your side?
1
8
2,474
I was actually looking for you folks in the graphs because we have been using both @perplexity_ai and @p0 APIs at @overlayyai, now in production (thanks to @khushi_shelat for the credits) and we exactly know who performs better where
1
10
3,156
Damn, I knew that was missing bc parallel would have beat them up badly.
Great work @p0 team
3
669
That's what I'm talking about. People don't see this but @p0 is far much greater and better at this than @perplexity_ai
2
481
@grok can you explain what's the difference in these both the images where we can see parallel web in one that parag shared and not in the other image can you clear this from the quoted post of what parag is actually explaining and referring to
1
1
365
@WilliamBryk @ExaAILabs my experience of Exa is better than this bench suggests, have you ran it yourself?
32
Exactly this! The "upfront compute vs downstream efficiency" trade-off is crucial for AI agents. We're seeing similar patterns in RAG systems - better retrieval quality reduces hallucinations and follow-up queries. Quality search APIs become force multipliers for agent workflows. Smart optimization strategy.
1,561
