Developer compares GPT-5.4 and mini pricing, advocates task-based routing

On November 10, 2026, the founder of PromptUnit published a post on DEV Community's AI section proposing a simple way to decide between GPT-5.4 and GPT-5.4 mini: route tasks by difficulty rather than defaulting to the larger model. Citing the OpenAI API changelog, the author notes that OpenAI released GPT-5.4 on March 5, 2026 and GPT-5.4 mini on March 17, 2026. Pricing for the two models is proportional across the board. GPT-5.4 mini is listed at $0.75 per million input tokens and $4.50 per million output tokens, while GPT-5.4 is listed at $2.50 and $15.00 respectively. Because 0.75 / 2.50 = 0.30 and 4.50 / 15.00 = 0.30, the mini variant costs exactly 30% of the flagship on both input and output. The practical implication drawn in the post is that any request mini is capable of handling can be served for 70% less. On the disclaimer side, the piece states that the author is the founder of PromptUnit, that it first appeared on the PromptUnit blog, and that the blog version is kept up to date. No benchmark or latency figures are included in the material, so the argument rests entirely on the API price list as quoted.