Last month, OpenAI began a limited preview of its GPT 5.6 series: Sol (flagship model), Terra (a balanced model for everyday work) and Luna (a fast and affordable model). Today, OpenAI announced that the GPT-5.6 Sol, along with the Terra and Luna models, will launch publicly tomorrow.
GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday.
We’re expanding preview access globally now. pic.twitter.com/Uk5HcfSc2e
— OpenAI (@OpenAI) July 8, 2026
Previously, as part of OpenAI’s engagement with the US government, the company started with a small preview of these models for a small group of trusted partners, and now the company will be expanding the preview access globally.
GPT-5.6 Sol, Terra & Luna
During evaluation, the GPT-5.6 Sol showed improved agentic capabilities in coding, biology, and cybersecurity, with additional safety and preparedness evaluations. With GPT-5.6, OpenAI has introduced a new Max reasoning effort and a new Ultra mode to reason deeply and accelerate complex work.
GPT-5.6 Sol, Terra and Luna models are revealed to demonstrate strong improvements in cyber capabilities. These models are developed with OpenAI’s most robust safeguards to date, with configurations matched to each model’s capabilities.
According to OpenAI, GPT-5.6 is trained to refuse prohibited cyber assistance, including when users attempt to disguise their intent or jailbreak the model. Real-time cyber and biology misuse classifiers provide another layer by evaluating output as it is generated. Flagged activity can also trigger account-level review across relevant conversations and risk signals, consistent with OpenAI’s terms and policies across content retention and review.
The company also used automated red-teaming and human red-teaming to find universal jailbreaks during the preview period.
The Sol, Luna and Terra names identify durable capability tiers that can advance on their own cadence.
Pricing
GPT-5.6 is priced per $1M tokens across three model sizes: Sol is $5 input/ $30 output, Terra is $2.50 input/ $15 output, and Luna is $1 input/ $6 output. For GPT-5.6 and later models, cache writes are billed at 1.25x the model’s unreached input rate, while cache reads continue to receive the 90% cached-input discount.
In July, OpenAI will launch GPT-5.6 Sol on Cerebras at up to 750 tokens per second. Access will initially be limited to select customers as the company expands capacity.
