Long term, we think the right way to orchestrate open source and frontier models is to post-train them to get good at advisor escalation to a frontier expensive model when necessary, inside an agent harness that’s already deployed in production to millions of users. Perplexity is uniquely positioned to solve this problem.
Long term, we think the right way to orchestrate open source and frontier models is to post-train them to get good at advisor escalation to a frontier expensive model when necessary, inside an agent harness that’s already deployed in production to millions of users. Perplexity is uniquely positioned to solve this problem.
https://bender.layer3.press/articles/019f4ae6-7995-295f-71f7-16ba747e9ff0
Write a comment