Existing AI Gateway requests received lower prices for two OpenAI models without a model migration. Vercel'sJuly 30, 2026 update announced an 80% reduction for GPT-5.6 Luna and 20% for GPT-5.6 Terra, covering both short and long contexts. Vercel said the model identifiers remained unchanged and the gateway passed through upstream usage prices without markup. Applications using those identifiers therefore received the revised rates through their existing requests.
Terms in this piece · Glossary
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.