OpenAI cut GPT-5.6 Luna by 80 percent and Terra by 20 percent, effective July 30. Luna now costs $0.20 per million input tokens and $1.20 per million output tokens, down from $1.00 and $6.00. Terra fell to $2.00 and $12.00. Sol did not move. [1]
Tuesday asked whether the Luna price cut still lands in shops. Wednesday is the same table. The cheap model got cheaper. The flagship grew a Fast mode at twice Sol's price for up to 2.5 times the speed. [1]
OpenAI says Luna now matches models that were frontier-class a year ago at roughly six cents on the dollar per task. It credits production-kernel work that cut serving cost about 20 percent and speculative-decoding gains above 15 percent. VentureBeat placed the move in a price war with Chinese and Google flash tiers. [1][2]
Trade notes repeated the table on July 31: Luna 80 percent, Terra 20, Sol unchanged, Fast at 2x. Subscription list prices for ChatGPT and Codex stay put; Luna and Terra simply burn fewer credits. AWS pricing was still rolling out when the company posted. [1][3]
Replit's Michele Catasta called Luna "the closest we've come to intelligence too cheap to meter." Ramp made Luna its default for background agents. Those quotes live on OpenAI's page. They are advertisements that happen to be specific. [1]
Tech wires moved on because a price list is not a third-day story. Buyers still test the 80 percent. Hard reasoning still costs Sol money. Volume work is what got the cut. Accessibility and quality are not the same product. Wednesday's market is still buying the first and pretending it bought the second.
The 80 percent is real. The quality story will be argued in evals, not in the price list. Cheap tokens do not make a careful model. They make a frequent one.
Notion said Terra matched GPT-5.5 quality at half the cost and 60 percent less time on scoped tasks. Those vendor sentences are how a cut lands: a default flipped, a credit burn slowed, a background agent pointed at Luna overnight. Finance teams will take a quarter. Developers already rerouted. [1]
Sol Fast replaces Priority Processing. Requests tagged priority roll into it. The honest caveat is in the table OpenAI did not discount. Hard reasoning still costs Sol money. Competitors still undercut Luna on raw tokens. What changed is that an OpenAI frontier-series name now sits in the bargain bin. Sol Fast, at a premium, is the refusal to fight it at the top. [1][2][3]
Wednesday's shops are the defaults, not the keynote. A background agent pointed at Luna is the landing. A finance committee that has not yet changed a SKU is not a contradiction. It is a calendar. The 80 percent waits for both.
A price cut that does not move Sol is a map of what OpenAI will not cheapen. Volume work got the discount. Judgment did not. That split is the product, not a footnote. [1]
-- THEO KAPLAN, San Francisco