Hello, world. We’re Principled Machines.
Language models are one form of intelligence. We’re building Cortex to combine them with algorithms, solvers, and reusable methods.
Our bet: better models make the system around them more powerful.
https://t.co/PIjM7J34Qz
Sharing an early result: Sierra verified Cortex at 73.2% on τ³-benchmark.
We build algorithmic complements to LLMs. We don’t fight what models do well. We fill in the rest.
https://t.co/MmTS0HVc3E