GPT-5.5 is priced at $5/1M input tokens and $30/1M output tokens, double GPT-5.4's pricing; GPT-5.5 Pro costs $30/1M input tokens and $180/1M output tokens
After months of rumors and reports that OpenAI was developing a new, more powerful AI large language model for use in ChatGPT …
Context & Ripple Effects
OpenAI had already raised standard GPT-5.4 API pricing above GPT-5.2, while offering Pro and Thinking variants and improved tool calling. GPT-5.5 extends that pricing progression: its standard rates are roughly twice GPT-5.4’s, but its Pro rates match the earlier GPT-5.4 Pro level.
Later GPT-5.6 pricing preserves the same $5/$30 top-tier rate while adding lower-priced Terra and Luna options, suggesting that OpenAI is building a price-and-capability ladder rather than moving every workload to a single premium model.
First-order effects
- API customers using GPT-5.5 face materially higher per-token costs than on GPT-5.4, especially for output-heavy applications.
- GPT-5.5 Pro is positioned as a premium option at $30 input and $180 output per million tokens; its unchanged price versus GPT-5.4 Pro makes the standard-tier increase the immediate commercial change.
Second-order effects
- Enterprise buyers and application developers will have stronger incentives to route routine or output-intensive work to cheaper models and reserve GPT-5.5 for tasks where its added capability justifies the spend.
- The price step increases the importance of usage monitoring and model-selection controls, because token-volume growth can raise bills faster when upgraded models are adopted by default.
Third-order effects
- If this tiering persists, model procurement will shift from choosing one frontier model to managing portfolios of models by task, latency, and cost.
- The durable competitive metric becomes cost per useful task rather than headline token price alone: vendors that can offer credible lower-cost tiers may gain leverage in budget-constrained deployments.
The trend: Frontier-model providers are moving toward segmented AI pricing, pairing premium capability tiers with lower-cost alternatives and pushing customers toward active model-routing economics.