OpenAI cuts GPT-5.6 Luna 80% and Terra 20% as the frontier price war reaches the flagship tier
On 30 July OpenAI cut GPT-5.6 Luna by 80% to $0.20 / $1.20 per million tokens and Terra by 20% to $2 / $12. A day later DeepSeek's V4-Flash-0731 landed at $0.14 / $0.28. The frontier is no longer competing only on capability — it is competing on price, and the cuts are arriving at the top of the lineup, not just the bottom.
An 80% cut is not a promotion; it is a repricing. Dropping Luna to $0.20 input signals that OpenAI now treats a large slice of frontier-tier inference as a commodity to be won on cost, not a premium to be defended on quality. When the lab that set the high end starts cutting the high end, the whole curve moves down with it.
DeepSeek shipping V4-Flash at $0.14 / $0.28 the very next day is the competitive mechanism made visible. The open-weight labs have been the price floor, and the closed labs are now cutting toward that floor rather than sitting above it. The gap between a premium API bill and a cheap-and-capable alternative is closing from both directions at once.
For buyers the consequence is that per-token cost is falling fast enough to change what is economically worth automating. Workloads that were too expensive to route through a frontier model six months ago cross the line as the price drops, which expands the market even as it compresses per-token margin — the classic shape of a commoditising input.
LLM-Stats — AI updates today (August 2026) — latest model releases → · LLM-Stats — LLM news today (August 2026) →