OpenAI: GPT-4o-mini Search Preview

OpenAI's GPT-4o-mini Search Preview, created on March 12, 2025, is a specialized model designed for web search within Chat Completions, featuring a substantial 128,000 context length. This model consistently performs among the fastest available, achieving an Infinityth percentile ranking across 8 benchmarks. It also offers highly competitive pricing, ranking in the Infinityth percentile for cost-efficiency across the same benchmarks. Demonstrating exceptional stability, the model boasts a 100% success rate across all 8 benchmarks, indicating outstanding reliability with minimal technical failures. In terms of specific performance, the model shows strong capabilities in General Knowledge (99.0% accuracy) and Ethics (99.0% accuracy), performing well above average. It also exhibits solid performance in Mathematics (84.0% accuracy) and Coding (80.0% accuracy). A notable strength is its cost-effectiveness across most benchmarks, frequently ranking in the 65th percentile or higher for price. However, a significant weakness is its performance in Instruction Following, where it scored 0.0% accuracy, suggesting a critical area for improvement. Its Hallucinations accuracy is also low at 28.0%, indicating a tendency to provide information rather than acknowledge uncertainty.