GPT-5 (medium) is a multimodal large language model developed by OpenAI, released in August 2025 as part of the GPT-5 flagship series. It is designed to balance high-level reasoning capabilities with operational efficiency, occupying a middle tier between the most computationally intensive variants and the lightweight "Mini" or "Nano" versions. The model supports both text and image inputs and features an adaptive reasoning system that allows it to adjust its internal chain-of-thought processing based on task complexity.
Built on a successor architecture to the GPT-4 series, the model is optimized for complex coding, scientific analysis, and agentic workflows. It incorporates a routing mechanism that dynamically selects between "fast" and "thinking" execution modes to maintain a balance of response speed and analytical depth. According to benchmark data, the medium configuration achieves performance parity with the highest-tier variants in many reasoning-heavy tasks while maintaining significantly higher throughput.
The model features a 400,000-token context window, which enables the processing of extensive documents and long-form conversational history. Its knowledge cutoff is established as September 2024. While the model is available through the OpenAI API with adjustable reasoning effort levels, it remains a proprietary system with no official parameter count disclosed.