Cerebras Systems and AI orchestration startup Callosum have partnered to integrate Cerebras’ wafer-scale AI accelerators into Callosum’s platform, which routes components of complex workflows to the most suitable combination of model and hardware in real time. Production benchmarks from the companies show a 4x performance improvement, a 70% reduction in compute costs and a 10% increase in task success rates versus running a single frontier model on standard infrastructure. Callosum, which announced a $100 million seed round on August 20, 2026, is developing an orchestration layer that functions like air traffic control for AI inference (the process of generating model outputs). Its compute strategy also includes Rebellions, Axelera, d-Matrix, Lumai and Tendrils, as well as Supermicro and HPE. Cerebras’ wafer-scale engine (a processor built across an entire silicon wafer) is reportedly 15 to 30 times faster than traditional GPU configurations for token generation and other latency-sensitive applications. The companies’ benchmarks focus on complex financial-services tasks, where lower inference costs and higher accuracy could have a significant effect on enterprise AI economics.