GPT-5.6 Terra

OpenAI's official model spec page (developers.openai.com/api/docs/models/gpt-5.6-terra) lists 'Reasoning: Higher' — the middle of a distinct three-way Highest/Higher/High ranking across the family (Sol=Highest, Terra=Higher, Luna=High), which does distinguish this tier unlike the identical 'Fast' speed label shared by all three. Rated 'moderate' relative to Sol's 'strong' (Highest) and Luna's 'weak' (High, the lowest of the three) — a within-family comparison, not an absolute judgment; OpenAI's own label for this tier is 'Higher', not 'moderate'.

OpenAI's GPT-5.6 launch page (openai.com/index/gpt-5-6/, required a live browser fetch — flat HTTP fetching was 403'd) states: 'GPT‑5.6 Sol is our best coding model yet. On the Artificial Analysis Coding Agent Index, GPT‑5.6 Sol with max reasoning sets a new state of the art at 80, 2.8 points above Fable 5... That advantage extends across the family: Terra performs just above Fable 5, while Luna outperforms Opus 4.8; each does so in roughly one-third of the time, with about half as many output tokens, and at approximately one-quarter the estimated cost.' This is a distinct, attributable per-tier claim (Terra vs. a named competitor, Claude Fable 5) via a third-party index OpenAI cites on its own page — independently verified by fetching the live page directly and confirming the exact quote. Rated 'moderate': a genuine (if narrow, 'just above') win over a frontier flagship-class competitor, but a smaller margin than Sol's outright 'new state of the art' claim.

OpenAI API docs (developers.openai.com/api/docs/models): supports Functions, Web search, File search, Computer use — same tool set as Sol and Luna.

1.05M-token context window (developers.openai.com/api/docs/models/gpt-5.6-terra) — same family-wide window as Sol and Luna.

OpenAI's own GPT-5.6 launch page (openai.com/index/gpt-5-6/, 'Availability and pricing' section, live-browser-verified 2026-07-20) explicitly calls Luna 'our fastest' model, implicitly placing Terra — the mid tier — between Luna and Sol. No direct speed adjective is given for Terra itself; this is a within-family relative rating derived from Luna's explicit claim, same relative-comparison approach already used for this doc's reasoning dimension.

OpenAI API docs: explicitly described as the 'GPT-5.6 model that balances intelligence and cost' — mid-priced at $2.50/$15 per 1M tokens, between Sol and Luna.