OpenAI GPT-6 Models Cut API Prices by 50%
OpenAI launched GPT-6 Sol and GPT-6 Luna, expanding its lower-cost AI API lineup based on GPT-6 Astra technology. GPT-6 Sol targets advanced coding and professional workloads, while GPT-6 Luna is designed for fast, large-scale everyday tasks.
GPT-6 Sol costs $2 per million input tokens and $10 per million output tokens. GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens. OpenAI says both models are about 50% cheaper than GPT-5.6 promotional rates and well below GPT-6 Astra’s pricing.
The new models also reportedly reduce coding deception and factual errors. Sol’s coding deception rate fell from 10.4% with GPT-5.6 to 1.3%, while Luna’s dropped from 9.5% to 2.8%. OpenAI reported stronger results in coding, computer use and professional-work benchmarks.
GPT-6 Sol and GPT-6 Luna are available through the API, ChatGPT Work and Codex for eligible paid and education subscribers. Free and Go users can initially test Luna through the desktop app. The GPT-6 launch could lower AI deployment costs, intensify competition across the AI infrastructure sector and increase demand for computing capacity. It has no direct effect on cryptocurrency prices, but traders may monitor related technology and semiconductor stocks.
Neutral
The news concerns OpenAI’s AI models and API pricing rather than a cryptocurrency or blockchain network. Therefore, it has no direct bullish or bearish catalyst for a specific crypto asset. In the short term, traders may shift attention toward AI-related equities, cloud providers and semiconductor companies, while crypto markets are likely to treat the announcement as neutral. Over the longer term, cheaper AI deployment could increase demand for computing infrastructure and potentially support blockchain projects associated with decentralised computing or AI, but the summaries provide no evidence of a direct token impact. Historical reactions to similar technology announcements typically depend on partnerships, revenue effects or token-specific adoption, none of which is confirmed here.