Sentient AI Unveils ROMA Open-Source Framework That Outperforms Google's Gemini in Complex Reasoning Tasks

Sentient AI releases ROMA, an open-source framework that breaks complex AI tasks into hierarchical trees, achieving 45.6% accuracy on reasoning benchmarks while outperforming Google's Gemini 2.5 Pro by more than double.

• Sentient AI releases ROMA (Recursive Open Meta-Agent), an open-source framework that structures AI agent workflows as hierarchical task trees where parent nodes break down complex goals into subtasks and aggregate results from child nodes
• ROMA operates through a minimal control loop of Atomize → Plan → Execute → Aggregate, enabling parallel execution of independent tasks while maintaining dependency ordering and providing full traceability across all node transitions
• ROMA Search, built on this framework, achieves 45.6% accuracy on SEALQA benchmarks, outperforming Kimi Researcher (36%) and Gemini 2.5 Pro (19.8%) in multi-source reasoning tasks