GPT-5.6 Luna Just Got 80% Cheaper — Here's What That Actually Means
Back to Blog
AI & TechTrendingGPT-5.6OpenAI

GPT-5.6 Luna Just Got 80% Cheaper — Here's What That Actually Means

Aug 9, 20269 min readClickWise Editorial

Three weeks. That's how long GPT-5.6 Luna held its launch price before OpenAI cut it by 80%. Companies don't slash a brand-new product from $6 to $1.20 because things are going well.

On July 30, Luna went from $1/$6 per million tokens to $0.20/$1.20. Terra dropped 20% too. Here's what forced OpenAI's hand, and how to actually benefit whether you write code or just pay for a chatbot.

The short version

  • Luna now costs $0.20 in / $1.20 out per million tokens
  • The cut came 3 weeks after launch — that's pressure, not charity
  • Terra fell 20%; flagship Sol didn't move
  • Your apps get smarter cheaper, even if you never touch an API

How much does GPT-5.6 Luna cost now?

As of July 30, 2026, GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens for short contexts — an 80% cut from its launch pricing of $1 and $6 just three weeks earlier. GPT-5.6 Terra dropped 20% to $2/$12, while the flagship Sol stayed at $5/$30.

GPT-5.6 Luna 80 percent price cut

From $6 to $1.20 per million output tokens in three weeks.

What actually changed

ModelOld price (in/out per 1M)New priceChange
GPT-5.6 Luna$1 / $6$0.20 / $1.20-80%
GPT-5.6 Terra$2.50 / $15$2 / $12-20%
GPT-5.6 Sol$5 / $30$5 / $30unchanged

To make that concrete: a chatbot handling 10 million output tokens a month — a busy customer-support bot, roughly — went from $60/month to $12/month on Luna. An app that was quoted $3,000/month in June now pencils out at $600. Entire product categories that didn't make economic sense eight weeks ago suddenly do.

Why OpenAI blinked

Nobody cuts a three-week-old product 80% voluntarily. The squeeze is coming from below: open-weight models like Kimi K3 and Qwen3.8-Max now deliver near-frontier quality at a fraction of the price, Google keeps undercutting on the cheap end, and Meta just gave away a capable agent model outright. When the models converge on quality, the only lever left is price.

There's a second story here too. OpenAI crossed one billion active users around the same announcement, and it's preparing the biggest IPO in history. Cheap Luna is a land grab: lock in a billion users and a million developers now, monetize the top tier later. Notice Sol's price didn't move — the hardest reasoning is where the margin lives.

How to actually take advantage

Depending on who you are
You build with the APIRe-route your traffic. Most requests don't need Terra or Sol — send summaries, extraction, and chat to Luna and keep the expensive tiers for the steps that fail on Luna. Teams that did this cut bills 60-80% in a week.
You pay for AI subscriptionsFalling API prices ripple into consumer apps. Free tiers get better because they now cost providers pennies. Before renewing anything, see our guide on which AI subscriptions are still worth it.
You were priced out beforeWhatever AI feature you shelved in 2025 because the math didn't work — redo the math. It's 5x cheaper now.
You're comparing providersDon't stop at OpenAI. The whole market is repricing; open models running locally cost you nothing but hardware.

💰 The 'short context' asterisk

The $0.20/$1.20 headline price applies to short contexts. Long-context requests bill differently, and output-heavy workloads still add up. If you're budgeting a product on Luna, price your real traffic pattern, not the press-release number.

And if $1.20 per million still sounds like too much, the logical endpoint of this trend is free: models you download and run yourself. We covered that path in how open models are killing AI subscriptions.

FAQ

How much does GPT-5.6 Luna cost now?+
As of July 30, 2026, GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens for short contexts — down 80% from the launch price of $1 and $6.
Why did OpenAI cut GPT-5.6 Luna's price?+
Competition. Open-weight models like Kimi K3 and Qwen3.8-Max deliver near-frontier quality at very low prices, and Google keeps undercutting on the low end. Cutting Luna 80% just three weeks after launch signals that frontier labs are losing pricing power.
Did GPT-5.6 Terra and Sol prices change too?+
Terra dropped 20% — to $2 input and $12 output per million tokens. The flagship GPT-5.6 Sol stayed unchanged at $5 input and $30 output.
Does the Luna price cut affect ChatGPT subscriptions?+
Not directly — ChatGPT Plus and Pro prices are unchanged. But cheaper API prices flow into the apps you use: expect more AI features in free tiers of other products, since developers now pay a fraction of what they did in June.

The bigger picture is that this cut isn't an event, it's a symptom. Intelligence is becoming a commodity, and commodities race to the bottom. We mapped the whole battlefield in the AI price war of 2026 — worth reading before you sign any annual AI contract.

Want more guides like this?

Join 50K+ readers getting weekly tips on AI, automation & making money online.

Subscribe Free
#GPT-5.6#OpenAI#AI Pricing#AI API#Luna#AI News 2026

Share this article