Wall St Engine(@wallstengine):Perplexity Search just took the top 3 spots on Artificial Analysis’ Search Index under the same model + agent setup: What stands out is that Perplexity stayed on top across compute levels, even when using less context, while competitors used more. Its low-context setting still beat every other search provider tested. It also had the lowest model inference cost per task at $0.028-$0.034, while all three context levels cost the same $5 per 1K searches. Medium looks like the sweet spot: best overall score at roughly $0.091 per task.

Wall St Engine

@wallstengine

Fast, accurate, consistent stock market news, earnings highlights & more. By Brillinsight. Not financial advice.

加入 March 2022

936 正在關注 187.7K 粉絲

Wall St Engine@wallstengine

2026.08.28 21:50

Perplexity Search just took the top 3 spots on Artificial Analysis’ Search Index under the same model + agent setup: What stands out is that Perplexity stayed on top across compute levels, even when using less context, while competitors used more. Its low-context setting still beat every other search provider tested. It also had the lowest model inference cost per task at $0.028-$0.034, while all three context levels cost the same $5 per 1K searches. Medium looks like the sweet spot: best overall score at roughly $0.091 per task.