OpenAI cuts GPT‑5.6 Luna price 80% and Terra 20%, adds Fast mode to Sol

OpenAI GPT‑5.6 Luna price cut GPT‑5.6 Terra reduction Fast mode Sol AI pricing competition

OpenAI announced on 30 July 2026 that it is lowering the API rates for two models in its GPT‑5.6 family. The Luna model, the smallest and fastest tier, saw its price cut by 80 % to $0.20 per million input tokens and $1.20 per million output tokens. The mid‑range Terra model was reduced by 20 % to $2 per million input tokens and $12 per million output tokens. The flagship Sol model’s price was left unchanged, but OpenAI introduced a new "Fast mode" that delivers up to 2.5× higher throughput for twice the standard processing rate (approximately $10 per million input tokens and $60 per million output tokens). Subscription fees for ChatGPT and Codex remain the same; however, token‑usage accounting for these plans has been adjusted so that the same subscription now yields more available work.

OpenAI said the reductions reflect efficiency gains across its model stack, inference infrastructure and routing, and were prompted by growing competition from lower‑cost Chinese AI models. The changes affect developers and enterprise customers using the OpenAI API and are intended to make large‑scale AI workloads more economical.

Entities

ARC-AGI-3 benchmark · Amazon Bedrock · ChatGPT · Fast mode · GPT-5.6 Luna · GPT-5.6 Sol · GPT-5.6 Terra · GPT‑5.6 Luna · GPT‑5.6 Sol · GPT‑5.6 Terra · Luna · OpenAI

Claims

What the coverage asserts, and how many sources carry each claim.

[● 17 SOURCES] OpenAI reduced GPT-5.6 Luna price by 80% to $0.20 per million input tokens and $1.20 per million output tokens. tecnoblog.net · www.blogspan.net · cryptobriefing.com · cryptonomist.ch · dev.classmethod.jp · +11 more
[● 12 SOURCES] OpenAI reduced the price of GPT-5.6 Terra by 20% to $2 per million input tokens and $12 per million output tokens. laverdadnoticias.com · tecnoblog.net · basic-tutorials.de · ai.cnmo.com · lajornadasanluis.com.mx · +5 more
[● 12 SOURCES] The price reductions were announced on 30 July 2026. laverdadnoticias.com · tecnoblog.net · basic-tutorials.de · ai.cnmo.com · lajornadasanluis.com.mx · +5 more
[● 2 SOURCES] OpenAI introduced a Fast mode for Sol that is up to 2.5× faster and costs double the standard rate ($10 per million input tokens, $60 per million output tokens). basic-tutorials.de · cryptobriefing.com
[● 2 SOURCES] ChatGPT and Codex subscription fees remain unchanged, but token usage limits are effectively increased for Luna and Terra due to the lower per‑token price. applealmond.com · cryptobriefing.com
[● 12 SOURCES] OpenAI reduced the price of GPT-5.6 Luna by 80% to $0.20 per million input tokens and $1.20 per million output tokens. laverdadnoticias.com · tecnoblog.net · basic-tutorials.de · ai.cnmo.com · lajornadasanluis.com.mx · +5 more
[● 12 SOURCES] OpenAI kept the price of GPT-5.6 Sol unchanged. laverdadnoticias.com · tecnoblog.net · basic-tutorials.de · ai.cnmo.com · lajornadasanluis.com.mx · +5 more
[○ 1 SOURCE] The new pricing is reflected on Amazon Bedrock as well as the OpenAI API. dev.classmethod.jp
[● 18 SOURCES] OpenAI said the price cuts were driven by efficiency improvements and competition from lower‑cost Chinese AI models. laverdadnoticias.com · tecnoblog.net · ai.cnmo.com · www.blogspan.net · www.iphonemod.net · +12 more
[○ 1 SOURCE] Amazon Bedrock aligned its pricing for GPT-5.6 Luna and Terra with OpenAI’s new rates on 30 July 2026. dev.classmethod.jp
[● 14 SOURCES] OpenAI said the reductions were driven by efficiency improvements and competition from lower‑cost Chinese AI models. laverdadnoticias.com · www.iphonemod.net · tecnoblog.net · www.1001web.fr · basic-tutorials.de · +9 more
[● 7 SOURCES] The price cuts apply to OpenAI API usage and increase the token quota for ChatGPT Work and Codex without changing subscription fees. www.iphonemod.net · basic-tutorials.de · lajornadasanluis.com.mx · www.blocktempo.com · cryptobriefing.com · +1 more