O3 Mini

O3 Mini is OpenAI's lighter-weight reasoning model, delivering strong reasoning at lower cost than full O3. Released January 31, 2025, it is optimized for science, mathematics, and coding challenges. The model exposes adjustable reasoning effort across three levels, low, medium, and high, so developers can tune how much computation each request spends: low for latency-sensitive tasks, medium for balanced performance, and high for harder problems that need deeper reasoning. O3 Mini supports structured outputs, function calling, and streaming, with a 200,000 token context window and up to 100,000 output tokens. It is text-only without vision, positioning it as an efficient option for reasoning workloads that do not need image input.