Supersonic Labs Releases Julia 1: A 144M-Parameter Open Decision Model That Runs on CPU

Supersonic Labs has open-sourced Julia 1, a 144.3 million parameter decision model explicitly designed to run on standard CPUs without requiring GPU hardware. The model targets structured decision-making tasks, enabling low-latency inference in environments where GPU access is unavailable, cost-prohibitive, or operationally impractical — such as edge devices, embedded systems, or budget-constrained cloud deployments. For developers building agentic pipelines, automation workflows, or decision-logic layers, Julia 1 offers a lightweight, open alternative to routing such tasks through larger hosted models. Its small parameter count makes it fast to load, cheap to run, and easy to fine-tune on domain-specific decision data. The open release means teams can inspect, modify, and self-host the model without API dependency.