GPT 5.6 Luna (Fast)

GPT-5.6 Luna is OpenAI's fast, cost-efficient model in the GPT-5.6 family, identified by the API model ID gpt-5.6-luna, and reached general availability on July 9, 2026 after a limited preview that began June 26. The page notes that the bare gpt-5.6 alias routes to Sol rather than Luna, which matters for developers pic For the GPT-5.6 Luna tier, the page documents a 1.05M-token context window with a 128K max output, a February 16, 2026 knowledge cutoff, and reasoning effort settings spanning none through max. Short-context pricing is listed at $1.00 per million input tokens and $6.00 per million output tokens, with cache write at $1.

CometAPI's explainer, updated September 2, 2026, frames GPT-5.6 as OpenAI's new model family announced on June 26, 2026 and available through a limited preview to selected trusted partners via the OpenAI API and Codex, with broader availability expected in ChatGPT, Codex, and the API in subsequent weeks. The family is The piece highlights Sol-specific capabilities that matter across the family: a new max reasoning effort level and an Ultra Mode that can spawn subagents to accelerate complex tasks, alongside gains in agentic coding, biology workflows, and cybersecurity, plus a refreshed safety stack. It notes that the rollout has bee

Big Hat Group's Codex Weekly roundup for the week of July 10, 2026, reports that GPT-5.6 reached general availability on July 9, 2026, after a two-week government vetting delay that kept it in limited preview since June 26 with about 20 partner organizations. The launch introduced a deliberate three-tier family (Sol, T Across the family, the "max" reasoning effort is available to all GPT-5.6 users in Codex, while "ultra" multi-agent acceleration is reserved for Plus and above in Codex and Pro/Enterprise in ChatGPT Work. Prompt caching now supports explicit cache breakpoints with a 30-minute minimum cache life, where cache writes bill

A June 26, 2026 AI-watch roundup reports that OpenAI kicked off a limited preview of the GPT-5.6 series that day, introducing a two-axis naming scheme where the number (5.6) denotes the generation and Sol, Terra, and Luna denote durable capability tiers that can evolve independently. The post quotes OpenAI announcing G Within the broader day-of news, the Luna tier is described as the most economical in the GPT-5.6 range, targeting high-volume workloads at the lowest cost in the lineup, complementing the new Max Reasoning Effort and Ultra Mode features introduced on Sol. The excerpted pricing table places GPT-5.6 Sol at $5/M input and