On July 9, 2026, OpenAI made the GPT-5.6 family generally available following a limited preview: Sol, the new flagship; Terra, a balanced model for everyday work; and Luna, the most cost-efficient tier. The release also formalized a new naming scheme in which the number identifies the model generation while Sol, Terra, and Luna are durable capability tiers that can advance on their own cadence.
OpenAI’s headline claims center on efficiency as much as raw capability. On Agents’ Last Exam, an evaluation of long-running professional workflows across 55 fields, GPT-5.6 Sol posted a new high of 53.6, which OpenAI says eclipses Claude Fable 5 by 13.1 points; the company also claims Terra and Luna outperform Fable 5 at around one-sixteenth the cost. A new “ultra” setting coordinates multiple agents across parallel workstreams for the most demanding tasks, and the API adds Programmatic Tool Calling plus a beta multi-agent mode that runs concurrent subagents inside a single request.
Availability spans ChatGPT, Codex, ChatGPT Work, and the OpenAI API, with a global rollout over 24 hours. Pricing per million tokens is $5 input / $30 output for Sol, $2.50 / $15 for Terra, and $1 / $6 for Luna. OpenAI also changed prompt caching economics: explicit cache breakpoints, a 30-minute minimum cache life, cache writes billed at 1.25x the uncached input rate, and a continued 90 percent discount on cache reads.
The launch landed in the middle of an unusually crowded two weeks, with xAI’s Grok 4.5 and Moonshot AI’s Kimi K3 arriving days later, all competing on price-performance rather than capability alone.