Multi-model agentic systems clearly are the future. Great post by cursor. Their new research shows that a frontier model as a planner and orchestrator with a workhorse cheaper model can materially lower the cost of total tokens on a project, resulting in a 15X cost improvement.
“Few moments in a large task genuinely require frontier intelligence, such as the original decomposition, the design decisions, and certain trade-offs. Once a frontier planner has collapsed the ambiguity into a detailed, explicit instruction, less expensive models simply have to follow it.”
This is increasingly becoming the core design pattern of complex agents, because the tokens used on large amounts of work on many tasks don’t require the same intelligence threshold as the planning step. By routing to different models based on where you are in the task, you get much greater efficiency overall.
This provides the template for where the applied layer in AI will differentiate. You can only drive this efficiency if you know the domain well and have the ability to work with multiple model tiers; companies that can do this will across coding, finance, legal, healthcare, life sciences, and other critical spaces will be in a strong position to get larger workloads that would otherwise be too expensive for the customer to deploy.
We briefly went insane and decided every software engineer should also become an expert in model benchmarks, thinking levels, and cache hit rates.
Cursor Router is out today. It picks the model for each task based on measured quality against cost. Fable 5 level performance for 60% less.
We had an incredible year @DevInterrupted - from breaking into Apple's Top 100 tech pods for the first time - to seeing 440% growth in our #SpotifyWrapped streams & recording our first in-person pods @ITRevDOES 🥳
So we're dreaming big - here are 8 ideas for Season 3 in Jan 🧵👇
Wow that Chris Rock/Will Smith moment just gave our Slack channel its first "Several people are typing at once" moment of this Oscars. I have no idea what happened after Jada's eyeroll.
INTERACT will have over $30B in engineering wisdom on display & we are hyped on how our speaker lineup is shaping up 🔥
We’re a month away, and so far we’ve announced ⬇️
🗣️ @SarvenazL, VP of Tech. at @AmericanExpress
🗣️ Dana Lawson, SVP of Eng. at @Netlify