OpenAI sharply cuts prices for GPT-5.6 Luna and Terra

OpenAI has cut the API prices for two of its GPT-5.6 models: Luna becomes 80 percent cheaper, Terra 20 percent cheaper.

As of July 30, GPT-5.6 Luna, the smallest and fastest model in the GPT-5.6 family, now costs 0.20 US dollars per million input tokens and 1.20 US dollars per million output tokens, down from 1 and 6 US dollars respectively. Terra, the mid-tier model in the lineup, dropped from 2.50 to 2 US dollars per million input tokens and from 15 to 12 US dollars per million output tokens.

The flagship model Sol remains unchanged at 5 US dollars per million input tokens and 30 US dollars per million output tokens. According to OpenAI, the savings stem from improved GPU kernels in the production environment as well as a more than 15 percent increase in token-generation efficiency, achieved in part by using the Sol model itself to optimize its own runtime efficiency.

According to OpenAI, the new prices also affect quota calculations in Codex and ChatGPT Work: tasks that use Luna or Terra going forward will consume a smaller share of the respective usage quota, while subscription prices themselves remain unchanged. In addition, OpenAI is switching the automatic code review feature Auto-review in the ChatGPT app and the Codex command line interface from GPT-5.4 to GPT-5.6 Luna, which the company says should cut the cost of this feature by around tenfold.

OpenAI: Faster but pricier mode for GPT-5.6 Sol

For its flagship model GPT-5.6 Sol, OpenAI also introduced a so-called Fast mode for API customers, which speeds up processing by up to 2.5 times without changing the model’s intelligence. The additional speed costs twice as much as standard processing and is aimed primarily at time-critical coding, research, and agentic tasks. For most other use cases, OpenAI says Fast mode is not necessary.

The price cut comes during a phase of intensifying price competition among leading AI providers: just a few days earlier, Anthropic had released its Claude Opus 5 model at the same price as its predecessor, Opus 4.8, while Google introduced Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, also new models designed for lower costs.

(red)