OpenAI Introduces GPT-5.6 Model Family in Phased Preview

OpenAI Unveils GPT-5.6 Model Family

OpenAI has announced the limited preview of its next-generation GPT-5.6 model family, including GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.6 Luna. The announcement, made on June 26, 2026, introduces Sol as the flagship frontier model, while Terra is presented as a balanced option for efficient daily tasks, and Luna as a fast and cost-effective choice for high-volume operations. This release follows approximately two months after the introduction of GPT-5.5.

Advanced Capabilities and Strategic Naming

The GPT-5.6 model family is designed to advance capabilities across several critical domains, including software engineering, computer use, professional knowledge work, scientific research, and cybersecurity. For its most capable model, GPT-5.6 Sol, OpenAI is introducing new reasoning effort options: a "max" mode to allow for deeper reasoning, and an "ultra" mode that leverages subagents to accelerate complex workflows beyond the capacity of a single agent. The company states that the GPT-5.6 models incorporate its most robust safeguards to date, with configurations tailored to each model’s specific capabilities, particularly in cybersecurity. OpenAI has also introduced a new naming convention with this release, drawing inspiration from celestial bodies (Sun, Earth, Moon) for Sol, Terra, and Luna. Under this system, the numerical designation, such as 5.6, identifies the model's generation, while Sol, Terra, and Luna denote durable capability tiers. These tiers are designed to evolve independently, providing developers and users with clearer distinctions based on intelligence, speed, and cost.

Limited Rollout and Access Parameters

The initial rollout of the GPT-5.6 family is a limited preview program, not a broad self-service offering. Access is currently restricted to a select group of trusted partners and organizations, available exclusively through the OpenAI API and Codex workspaces. Individual consumers are not included, and there is no public application process or waitlist. OpenAI plans to make the models generally available in ChatGPT, Codex, and the API in the coming weeks. This phased release strategy stems from ongoing engagement with the U.S. government. OpenAI previewed its plans and the models' capabilities with the government prior to launch and is beginning with a limited group of partners whose participation details have been shared with government officials. This approach allows for additional testing and coordination, aiming to enhance safety and usability before a broader release. Access approvals are specific; an organization's approval for the API does not automatically extend to Codex, and vice versa.

Pricing Structure for New Models

The GPT-5.6 models are priced per 1 million tokens, with varying rates for input and output across the three tiers: - GPT-5.6 Sol is priced at $5 for input tokens and $30 for output tokens. - GPT-5.6 Terra is priced at $2.50 for input tokens and $15 for output tokens. - GPT-5.6 Luna is the most cost-efficient, priced at $1 for input tokens and $6 for output tokens. The strategic, government-coordinated limited release of OpenAI's GPT-5.6 family underscores a growing emphasis on safety and responsible AI deployment, which will shape future accessibility for developers and users.