OpenAI: GPT-5.4

OpenAI's GPT-5.4 is a commercial model that accepts text, image, and file inputs, with a context window of 1,050,000 tokens and up to 128,000 output tokens per response. It supports tool use and reasoning, which suits agent workflows and multi-step tasks over large documents or codebases. Structured output support and weight availability are not listed, so teams that depend on either should verify them before committing. Pricing is $2.50 per million input tokens and $15.00 per million output tokens. The output rate makes long generations the main cost driver, while the large context keeps input-heavy work comparatively affordable. Its blended benchmark score of 97.0 rests on a single independent benchmark, EQ-Bench, where it scored 75.6. That is thin evidence for broad claims about quality. Shortlist it if you need very long context, multimodal input, and tool support, but run your own evaluations before treating its standing as proven.