Advancing the price-performance frontier with GPT-5.6

What changed

Source summary

OpenAI cut API pricing for GPT-5.6 Luna by 80% and GPT-5.6 Terra by 20%, and introduced a Fast mode for GPT-5.6 Sol offering up to 2.5x faster responses at double the price, effective 30 July 2026.

Why it matters to your work

Practical interpretation

If your business uses GPT-5.6 via the API for high-volume tasks, this is a straightforward cost cut — Luna and Terra now cost significantly less to run at scale, so routine automation or bulk content workflows get cheaper without changes on your end. If speed matters more than cost for a task, the new Fast mode is worth turning on for GPT-5.6 Sol.

When this could help

Relevant when reviewing API cost for GPT-5.6 workloads, or choosing between Standard and the new Fast mode for latency-sensitive requests.

OpenAI cut API pricing for GPT-5.6 Luna by 80% (to $0.20/million input tokens, $1.20/million output tokens) and Terra by 20% (to $2.00/million input tokens, $12.00/million output tokens), effective 30 July 2026, with no pricing change for Sol. A new Fast mode replaces Priority Processing in the API, running up to 2.5x faster than standard processing at twice the price with no change in intelligence. In ChatGPT Work and Codex, Terra and Luna now consume fewer credits per use, though subscription prices and quota budgets are unchanged.

What to check

• Prior per-token prices for Terra and Luna are not published — only the percentage cuts and new absolute prices are stated, so an exact before/after comparison is not possible from this source. Source ↗
• Performance claims (e.g. Luna comparable to frontier-class models from a year ago, nearly 99% lower cost per task on Agents Last Exam) are OpenAI own relative comparisons against older models, not independently verified fixed benchmarks. Source ↗