Rizzo AI Academy introduced the local engine Rizzo Flow

Developers from Rizzo Flow have released an open-source tool of the same name for local execution of API requests of the Jev type. The solution is designed to work on the user's own equipment, ensuring complete data privacy without accessing external cloud servers. The main task of the system is to generate typed solutions with probability indications instead of standard text output. The tool supports a compatible HTTP interface, which allows it to be integrated into existing data processing pipelines. The project is distributed under a free license, allowing the deployment of decision-making infrastructure on local resources. The technical implementation of Rizzo Flow focuses on strict typing of responses, which is critical for automated systems that require predictable data formats. Unlike classic LLMs that output unstructured text, this engine returns structured probabilistic indicators, simplifying further programmatic processing of results. Using local launch eliminates delays associated with network requests to API of third-party providers. The tool is designed for developers creating autonomous systems where high accuracy and control over the decision-making process are required. The transition to local API implementations allows reducing operational expenses for using cloud tokens and provides independence from the update policy of external services.