Robert Ta shows how Cursor's agent swarm rebuilt SQLite in Rust using a planner-worker architecture. Mixing Fable-5 as the planner with cheaper workers like Opus 4.8 cut costs to $13.39 while passing all 835 test cases with 6.5x less code than the previous version.
Cursor had an agent swarm implement SQLite in Rust from scratch against an 835-page manual graded by SQL Logic Test. The cost to rebuild the SQLite by model mix, you could see Fable 5 here with the highest cost, $20,000. Lowest cost, Opus 4.8 plus Composer 2.5, where they split up the workloads to rebuild SQLite. That came in at almost at elite score. Actually $13.39. So only $2 off of LEET. The architecture was planner agents being the frontier models. They decompose goals. Worker agents, fast, cheap models execute on them. And Wilson Lin says a planner doesn't implement, so its context never fills with low-level detail. Worker never plans, so it can spend all its context on one narrow piece of work. And that was the architecture they went for. They saw that GPT-5.5 alone burned 14.7 billion tokens and Fable-5 12.2 billion tokens. And the Frontier as planner plus cheap worker mix delivered near identical quality at 7.9x lower cost than Frontier only. Quote, once a Frontier planner has collapsed the ambiguity into a detailed explicit instruction, less expensive models simply have to follow it. And the quality bar— the new swarm hit 100% test pass on the Fable-5 mix using 9,908 lines of code versus 6 4,000 lines of code for the old Swarm, which is about 6.5x less code. So the thesis is really that frontier for planning and orchestration, and then lower-level models that cost less for the actual implementation.
Personally, in, in my work, I've been using, uh, Fable-5 as a planner, and then, uh, GPT-5.6 SOL is like the implementation. I know it's not the cheapest way to get it done, but for unbound planning, especially creative planning, Fable is like a really great model, and then it can either orchestrate other agents or, or run them. So here they showed their own Composer 2.5 working with Opus 4.8. That was the cheapest one, right? And I think Opus 4.8 on a bounded task, like really create something with a fixed spec, is actually a pretty good model to do that. And Fable 5 is kind of like overkill. There's not enough creativity needed where Fable can really show its value. And I think you can see that I would use Fable in something where it could produce a better result than Opus 4.8. And expect to see results there. I was curious they didn't try with just Cursor 2.5 alone. I wonder what's up with that. Like if it just isn't well suited for sort of like the planning and orchestration tools. But, uh, yeah, these are exactly the kind of experiments you're kind of like running every day as a builder right now. Right. And they just did it at scale and put some numbers to it. It's really helpful for the rest of us.
