OpenAI has introduced GPT-5.4 mini and nano, compact versions of its flagship model designed for speed and lower costs. While mini targets coding and reasoning tasks, nano focuses on lightweight workloads like classification and data extraction. Both are available via API, with mini also accessible in ChatGPT.

OpenAI has launched GPT-5.4 mini and GPT-5.4 nano, its most capable small models to date, designed to bring the strengths of its flagship GPT-5.4 to high-volume, latency-sensitive workloads at a fraction of the cost.

What Are GPT-5.4 Mini and Nano?

These are compact, highly efficient versions of OpenAI's GPT-5.4 model, optimised for speed and cost rather than maximum capability. Think of them as the workhorses of the GPT-5.4 family - not built to replace the full model, but to handle the fast, repetitive, and supporting tasks that make up the bulk of real-world AI workflows.

GPT-5.4 mini is a significant upgrade over its predecessor, GPT-5 mini, with improvements across coding, reasoning, multimodal understanding, and tool use - while running more than 2x faster. In benchmarks, it approaches GPT-5.4-level pass rates at considerably lower latency, making it one of the strongest performance-per-speed tradeoffs available for coding workflows.