OpenAI's new mid-tier GPT-6.1-family model — 50% cheaper than both GPT-5.6 Terra and GPT-6.1 Sol, built for computer-use, visual reasoning, and coding.
GPT-6.1 Terra bills $1 / 1M input and $5 / 1M output for prompts up to 272K tokens; longer prompts are billed at $2 / $7.50 per 1M. Cached input reads are $0.05 / 1M (≤272K) or $0.10 / 1M (>272K) — 5% of the input rate.
OpenAI released GPT-6.1 Terra on October 26, 2026, filling out the GPT-6.1 family between GPT-6.1 Sol and GPT-6 Luna. It's priced 50% below both GPT-6.1 Sol and its own predecessor-generation namesake, GPT-5.6 Terra, at $1 input / $5 output per 1M tokens — OpenAI's cheapest model above the Luna budget tier.
OpenAI positions Terra for computer-use, visual-reasoning, and coding workloads that don't need Sol's higher ceiling. It's available in ChatGPT Work and Codex on Plus, Pro, Enterprise, and Edu plans, and via the API as gpt-6.1-terra.
GPT-5.6 Terra remains available separately at its existing $2/$12 price — OpenAI has not announced a retirement date for it alongside this launch.
Terra sits as the value pick in the GPT-6.1 lineup: strong coding, vision, and computer-use performance at a fraction of Sol's price, for teams that don't need Astra- or Sol-level reasoning depth.
| Model | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| GPT-6.1 Terra | $1 | $5 | 1.05M |
| GPT-6.1 Sol | $2 | $10 | 1M |
| GPT-6 Luna | $0.10 | $0.50 | 1M |
| GPT-5.6 Terra | $2 | $12 | 1M |
| Claude Haiku 5.5 | $0.10 | $0.50 | 1M |
GPT-6.1 Terra undercuts its own predecessor GPT-5.6 Terra by half and sits well below GPT-6.1 Sol, but it's still a step up in price from the budget-tier GPT-6 Luna and Claude Haiku 5.5 — the trade-off being Terra's larger context window and stronger coding/vision performance.