OpenAI cuts GPT-5.6 Luna price by 80% just weeks after launch

OpenAI on Thursday cut prices for two of its three GPT-5.6 models just three weeks after their public launch, reducing the cost of its cheapest tier by 80% and its mid-range offering by 20% as the company faces mounting pressure from cost-conscious enterprise customers and cheaper Chinese rivals.axios+1

The price of GPT-5.6 Luna, the fastest model in the family, drops to 20 cents per million input tokens and $1.20 per million output tokens, down from $1 and $6. GPT-5.6 Terra falls to $2 and $12 per million input and output tokens, respectively, from $2.50 and $15. The flagship Sol model remains unchanged at $5 input and $30 output.unite+1

Efficiency Gains Enable the Cuts

OpenAI attributed the reductions to infrastructure optimizations detailed in an engineering post published July 29. The company said improvements including speculative decoding, fine-grained load balancing, cache optimization, and custom GPU kernels cut end-to-end serving costs by 20% and raised token-generation efficiency by more than 15%.news.aibase+1

"Our strategy remains focused on advancing both capability and efficiency so each generation of intelligence can accomplish more work at a lower cost," OpenAI said in a release.cnbc

Competitive Pressure Mounts

The repricing arrives amid intensifying competition across the AI industry. Chinese startup Moonshot AI released its open-weight Kimi K3 model earlier this month, outperforming some leading American models on industry benchmarks. Anthropic responded by releasing Claude Opus 5, while Microsoft ↗ 3.02% CEO Satya Nadella highlighted cost-effective models during the company's earnings call on Wednesday. Google Alphabet Inc. ↗ 6.88% also debuted three new models this month aimed at undercutting competitors on cost.cnbc

At its new price point, Luna undercuts Anthropic's cheapest published model by a factor of roughly five on input tokens, according to Reuters. Terra now sits below the $3 per million input tokens that Anthropic charges for its mid-tier Claude Sonnet offering.globalbankingandfinance+1

The Broader Cost Reckoning

The cuts reflect a broader shift in enterprise AI spending. After an era of aggressive adoption where companies encouraged employees to use AI freely, many firms are now actively working to rein in ballooning bills. Amazon's Amazon.com, Inc. ↗ 15.32% engineering organization moved to cap AI spending after cost overruns, and OpenAI itself shipped hard spend limits for API customers on July 22.unite+1

Analysts have noted that while lower prices could boost usage, they may also strain finances for OpenAI and Anthropic ahead of anticipated initial public offerings. As Axios reported, the speed of the cuts — just three weeks after launch — is unusual in an industry where price reductions typically come months after a model ships.globalbankingandfinance+1