The GPT-5.6 record is concrete now, but still narrow. OpenAI has started a limited preview of a three-model family: Sol, Terra, and Luna.
That is worth tracking. It is not a model-picker launch for ordinary ChatGPT users.
Current status
The help-center record is the anchor. OpenAI says the preview is available through the OpenAI API and Codex for a limited group of trusted partners and organizations. The same record says GPT-5.6 is not available in ChatGPT during the preview, there is no public application or waitlist, and no general-availability date has been announced.
That leaves Hugin with a simple status label:
- Confirmed: GPT-5.6 exists as an official limited preview.
- Confirmed: Sol, Terra, and Luna are the named models in the family.
- Confirmed: Preview access can include API, Codex, or both, depending on the approved organization and workspace.
- Not confirmed: a public self-service signup, public waitlist, or broad ChatGPT rollout.
- Not confirmed: a dated general-availability launch.
The three lanes
The family is easier to understand as three lanes, not one monolithic launch:
- Sol is the flagship lane: the strongest model in the family, aimed at hard coding, professional work, scientific research, and cyber workflows.
- Terra is the cost-control lane: still strong, but priced below Sol.
- Luna is the speed lane: the fastest and most cost-efficient option in the family.
The pricing receipt is already public in the help article. gpt-5.6-sol is
listed at $5 per 1M input tokens and $30 per 1M output tokens. gpt-5.6-terra
is listed at $2.50 input and $15 output. gpt-5.6-luna is listed at $1 input
and $6 output. The same page says GPT-5.6 adds explicit cache breakpoints and a
30-minute minimum cache life, with cache writes billed above uncached input and
cache reads keeping the cached-input discount.
Why this tracker matters
The interesting part is not just "new model." It is the release shape.
OpenAI's launch post introduces a max reasoning effort and an ultra mode
that uses subagents for complex work. That moves the public question away from
"is it bigger?" and toward "where is it available, how is work reviewed, which
lane should a team use, and what proof exists after a long task finishes?"
The system card is the second anchor. It describes the safeguards around Sol, Terra, and Luna, especially for biological, chemical, and cybersecurity risk. That matters because the access story and the safety story are moving together: broader rollout should be judged only when access language, release notes, pricing, status pages, and safety records line up.
What Hugin should watch next
The next useful receipts are practical:
- A dated OpenAI post or help-center update that changes "limited preview" into broader availability.
- API documentation showing the model IDs as generally usable outside selected organizations.
- Codex release notes showing which workspaces or plans receive Sol, Terra, or Luna.
- ChatGPT release notes showing whether GPT-5.6 enters the consumer model picker.
- Pricing-page updates that match or revise the help-center prices.
- Status-history notes if availability, safeguards, or preview access create a visible service issue.
Until those receipts appear, the honest headline is simple: GPT-5.6 is public enough to track, but not public enough to call a general rollout.