OpenAI: o3 Mini High

OpenAI o3-mini-high, released on February 12, 2025, is a specialized language model designed for STEM reasoning, building upon the o3-mini model with enhanced reasoning effort. The model exhibits moderate speed performance, ranking in the 28th percentile, and is positioned at premium pricing levels, falling into the 13th percentile for cost-efficiency. A significant strength is its exceptional reliability, achieving a 100% success rate across all benchmarks, indicating consistent and usable responses. The model demonstrates strong performance in core STEM areas, achieving perfect accuracy in General Knowledge and Reasoning benchmarks, and a high 92.0% in Mathematics and 90.0% in Coding. Notably, it is the most accurate model at its price point and among models of similar speed for General Knowledge and Reasoning. While excelling in these domains, its performance in Instruction Following (62.0% accuracy) and Ethics (98.0% accuracy) is less outstanding compared to its STEM capabilities. Hallucinations are well-managed with 94.0% accuracy, indicating a good ability to acknowledge uncertainty. Its primary strengths lie in its robust reasoning capabilities and reliability, making it a strong contender for applications requiring precise, error-free outputs in technical fields, despite its premium cost and moderate speed.