Skip to content
TopicTracker
From simonwillison.netView original ↗
TranslationTranslation

Quoting OpenAI

OpenAI announced a limited preview of the GPT-5.6 series, introducing three models: Sol (flagship), Terra (balanced for everyday work), and Luna (fast and affordable). Pricing ranges from $1 to $5 per million input tokens and $6 to $30 for output. The preview begins with a small group of trusted partners, with broader release planned in the coming weeks after coordination with the U.S. government.

Background

OpenAI has announced GPT‑5.6 ("Sol"), its latest large language model, alongside two smaller/cheaper variants called Terra and Luna. Key context: this follows a long pattern of incremental model releases (GPT‑4, GPT‑4o, GPT‑4.1, GPT‑5, GPT‑5.5, now 5.6). Each new version typically improves reasoning, reduces cost, or both. The pricing (per "token," roughly 0.75 words) shows a three-tier strategy: a powerful flagship (Sol), a balanced workhorse (Terra), and a cheap fast model (Luna). "Prompt caching" — a technique that stores repeated input to avoid recomputing it — gets more predictable billing. The mention of U.S. government preview and "limited preview with trusted partners" reflects growing regulatory scrutiny of advanced AI models, where companies coordinate with government agencies (e.g., the White House AI Executive Order) before wide release. Simon Willison's blog often catalogs and links to official AI announcements for readers who follow the field closely.

Related stories

  • Summary of METR's predeployment evaluation of GPT-5.6 Sol

    10.0

    METR conducted a predeployment evaluation of GPT-5.6 Sol, assessing its capabilities and risks before release. The evaluation focused on the model's performance in areas such as autonomous task completion and potential misuse. Findings informed safety decisions ahead of deployment.

  • OpenAI Announces, But Is Blocked From Releasing, New GPT-5.6 Models

    7.0

    OpenAI announced a limited preview of its GPT‑5.6 series but is releasing them gradually because the U.S. government requested a staggered rollout, with customer-by-customer approval. Commerce Secretary Howard Lutnick later cautioned against launching without broader agency sign-offs.

  • Assessing GPT-5.6 Sol Against Cybersecurity Benchmarks

    3.0

    OpenAI's GPT-5.6 Sol was evaluated against standard cybersecurity benchmarks, showing improved performance in vulnerability detection and threat analysis compared to previous models, though it still struggles with highly specialized attack vectors and adversarial inputs.