Re:SWE-in-a-team: a coding benchmark for software factories
by Marcuss2 on Tuesday September 01, 2026 @12:19PM
(#49524112)
Replace all of the engineers with open weight models. And add GLM 5.3 Flash as a builder. I am interested how will they fare.
Re:SWE-in-a-team: a coding benchmark for software factories
by onder_ceylan on Tuesday September 01, 2026 @01:57PM
(#49525422)
Yes, curious. Will do!
I'm interested to see the total inference cost of a ticket type, that's why total cost is compared. Even though most of the cost is there in other stages, there's still huge difference even with one variable configuration in the loop.
The cost of each builder is also shown in mission breakdown, but good feedback for extending some of the tables, thanks!