GPT‑5.5 API set at roughly 2× prior rates, prompting backlash over pricing
- OpenAI added GPT‑5.5 to its API on April 24, listing standard prices of $5 per million input tokens and $30 per million output tokens for the new flagship model. - Those rates are about double GPT‑5.4’s base prices on OpenAI’s pricing page, while GPT‑5.5 Pro stayed far higher at $30 input and $180 output per million tokens. - OpenAI says GPT‑5.5 uses fewer tokens on some coding tasks, but developers are weighing higher list prices against any efficiency gains. (openai.com)
OpenAI put GPT‑5.5 into its API on April 24 and priced the flagship model above GPT‑5.4, sharpening a cost debate that started when the model launched in ChatGPT on April 23. (openai.com 1) (openai.com 2) On OpenAI’s pricing page, GPT‑5.5 is listed at $5 per 1 million input tokens, $0.50 per 1 million cached input tokens, and $30 per 1 million output tokens. GPT‑5.4 is listed at $2.50 input, $0.25 cached input, and $15 output. (openai.com) That makes GPT‑5.5 roughly 2 times GPT‑5.4 on the base token rates shown publicly by OpenAI. GPT‑5.5 Pro is listed separately at $30 per 1 million input tokens and $180 per 1 million output tokens. (openai.com 1) (openai.com 2) The pricing matters because application developers do not buy a flat subscription when they use the application programming interface, or API. They pay by token, a usage unit that counts pieces of text sent in and generated back. (openai.com 1) (openai.com 2) A higher token price does not always mean a higher final bill. In its GPT‑5.5 launch post, OpenAI said the model “uses significantly fewer tokens” to complete the same Codex tasks and matches GPT‑5.4 on per‑token latency. (openai.com) That tradeoff is the center of the argument. If GPT‑5.5 finishes a coding or research job in far fewer steps, some teams could absorb the higher list price; if it does not, the posted rates flow straight into inference costs. (openai.com) (openai.com) OpenAI described GPT‑5.5 as its “smartest and most intuitive to use model yet,” with gains in agentic coding, computer use, knowledge work, and early scientific research. The company said nearly 200 early‑access partners tested real use cases before release. (openai.com) The company also rolled the model out across paid ChatGPT tiers and Codex before adding the API a day later. OpenAI said GPT‑5.5 reached Plus, Pro, Business, and Enterprise users in ChatGPT, while GPT‑5.5 Pro reached Pro, Business, and Enterprise. (openai.com) The immediate effect is that OpenAI now shows a clearer pricing ladder across its current lineup, from lower‑cost GPT‑4.1 variants up to GPT‑5.5 and GPT‑5.5 Pro. That gives buyers a more explicit menu for routing simple tasks to cheaper models and reserving the flagship for harder work. (openai.com) For developers, the new question is less whether GPT‑5.5 costs more on paper than whether its claimed efficiency offsets that increase in production. OpenAI’s own pricing page now makes that comparison unavoidable. (openai.com) (openai.com)
