OpenAI: GPT-5.4 Mini

GPT-5.4 Mini is OpenAI's smaller-footprint model that accepts text, images, and files as input, with a 400,000-token context window and a 128,000-token output limit. It supports tool use and reasoning, making it capable of multi-step and agentic workflows, though structured output support is unconfirmed from available data. At $0.75 per million input tokens and $4.50 per million output tokens, it sits in the budget-to-mid tier of capable models. Its blended benchmark score of 70.8 across three benchmarks is modest, and that limited coverage means the picture is not yet complete. The standout figure is its coding score of 92.5, which makes it worth shortlisting for development-heavy workloads where cost discipline matters. Its agentic score of 49.8 is relatively weak, so teams building complex autonomous pipelines should compare it carefully against alternatives before committing.