- OpenAI launched GPT-5.5 with agentic coding, computer use, and scientific research capabilities.
- API pricing doubles to $5/$30 per 1M tokens, with Pro tier at $30/$180 per 1M tokens.
- The model will power OpenAI’s upcoming ‘super app’—and it’s rolling out to Plus and Pro users now.
OpenAI has officially introduced GPT-5.5, a model the company claims is its most capable yet when it comes to handling complex tasks without much hand-holding. The launch comes as the AI wars continue to heat up—Anthropic just overtook OpenAI in secondary market valuations.
According to OpenAI, GPT-5.5 is priced significantly higher than its predecessor—$5 per 1 million input tokens and $30 per 1 million output tokens, double what OpenAI charged for GPT-5.4. The Pro tier is even steeper at $30 per 1 million input tokens and $180 per 1 million output tokens. The company says the premium is justified by what it calls “significant improvements” in agentic coding, computer use, and early-stage scientific research.
OpenAI isn’t just launching a model—it’s building a whole new interface around it. The company confirmed GPT-5.5 will power its upcoming “super app,” a bid to move beyond the chatbot format toward something that can actually run entire workflows. Bloomberg’s Rachel Metz first reported the details this afternoon.
Why GPT-5.5 Costs More—and Who Gets Access First
The pricing jump reflects OpenAI’s bet that developers and enterprises will pay for quality over quantity. Unlike GPT-5.4, which was positioned as a general-purpose workhorse, GPT-5.5 is designed for workloads that require reasoning across longer contexts—think multi-step code refactoring, autonomous browser automation, or hypothesis testing in research pipelines.
ChatGPT Plus, Pro, Business, and Enterprise subscribers get immediate access in both the main chat interface and Codex, OpenAI’s programming assistant. GPT-5.5 Pro is reserved for Pro-tier customers and above on the ChatGPT side—a clear segmentation play meant to push power users up the pricing ladder.
The model’s rollout timing is aggressive: it launched the same day it was announced, which is unusually fast even for OpenAI’s standards. Axios noted that OpenAI has been unusually tight-lipped about model specifications, declining to publish benchmark scores or safety evaluations alongside the launch.
What hasn’t changed is OpenAI’s appetite for selling shovels in the AI gold rush. The company is effectively betting that even as AI prices fall industry-wide—witness the recent price cuts from Anthropic and Google—there’s still a healthy market for premium capability. Whether developers agree will become clear within weeks as API usage data rolls in. GPT-5.5 is live now.