OpenAI released its GPT-5.6 model suite on Friday, just one day after the company agreed to stagger the launch at the request of the Trump administration. The move comes amid growing tension between the White House and the AI industry over regulatory oversight.

The new suite includes three models: Sol, the flagship; Terra, a mid-tier option for high-volume workloads; and Luna, a fast and affordable everyday model. OpenAI says the models excel at coding, cybersecurity and biology, and maintain focus during long-horizon agentic AI tasks.

Three-Model Tier Targets Different Needs

Each model is priced based on capability and intended use case. Sol, the most powerful, is priced at $5 per million input tokens and $30 per million output tokens. That is roughly half the input cost of Anthropic's Claude Fable 5, which charges $10 input and $50 output per million tokens. Terra and Luna offer lower pricing tiers for developers who need efficiency or scale.

  • Sol: Flagship model for advanced reasoning, coding and agentic tasks. Highest cost per token.
  • Terra: Medium-tier model optimized for high-volume work and balanced performance.
  • Luna: Fast and affordable model for everyday queries and low-latency applications.

Pricing War and Competitive Pressures

OpenAI's pricing strategy signals an intensifying battle with rivals like Anthropic and Google. By undercutting Claude Fable 5 on input costs, OpenAI aims to attract developers who are sensitive to API expenses. However, Sol's output cost is still relatively high, which may push users toward the cheaper Terra or Luna models for production workloads. The company is betting that its specialized strengths in cybersecurity and biology will differentiate the suite from competitors that focus on general-purpose reasoning.

Why This Matters

The timing of the release is as significant as the technology. The Trump administration's request to delay the rollout highlights a growing rift between Silicon Valley and federal regulators. Developers and enterprises who rely on OpenAI's models now face an unpredictable regulatory environment that could affect future product availability and pricing. At the same time, the rapid pace of model releases suggests that AI companies are racing to lock in market share before new rules take shape. Users should expect more iterative releases and tighter integration with agentic AI systems in the coming months.