AIMay 1, 2025Microsoft AI Released Phi-4-Reasoning: A 14B Parameter Open-Weight Reasoning Model that Achieves Strong Performance on Complex Reasoning Tasks
