pricing gpt 5.6 luna modle

Question

pricing gpt 5.6 luna modle

Verma, Anand 5

if I purchase 5.6 luna from openai the pricing is been reduced 80%
when can we see the same here is azure

Karnam Venkata Rajeswari 5,170 Reputation points Microsoft External Staff Moderator

2026-08-11T19:08:39.42+00:00
Hello @Verma, Anand ,

Welcome to Microsoft Q&A .Thank you for reaching out to us.

Effective August 1, 2026, Azure OpenAI customers receive the same updated price reductions for GPT-5.6 Luna and Terra announced by OpenAI. Microsoft has confirmed that Azure is committed to pricing parity with OpenAI for applicable Azure OpenAI offerings.

For GPT-5.6 Luna on Standard Global (short context), the updated rate is $0.20 per 1M input tokens, $0.02 per 1M cached input tokens, $0.25 per 1M cached writes, and $1.20 per 1M output tokens.

If your deployment still displays the previous $1.00/$6.00 rates, we recommend checking the actual charges in Azure Cost Management or your invoice, rather than relying only on the public pricing page. Microsoft provides guidance on monitoring Foundry costs here:

Also, please verify the deployment type. Global Standard, Data Zone, Priority Processing, and Provisioned Throughput have different pricing models, so the rates may not be directly comparable.

If your actual billing still shows the old rates for usage after August 1, 2026, Please reach out to use by opening an Azure Support billing request for further assistance

The following references might be helpful , please check them out

Microsoft announcement — GPT-5.6 now available in Microsoft Foundry

Microsoft Foundry deployment types and pricing guidance

Plan and manage costs for Microsoft Foundry

How to create an Azure support request - Azure portal | Microsoft Learn

Thank you

2 answers

Your answer

Karnam Venkata Rajeswari 5,170 Reputation points Microsoft External Staff Moderator

2026-08-11T19:08:39.42+00:00

Hello @Verma, Anand ,

Welcome to Microsoft Q&A .Thank you for reaching out to us.

Effective August 1, 2026, Azure OpenAI customers receive the same updated price reductions for GPT-5.6 Luna and Terra announced by OpenAI. Microsoft has confirmed that Azure is committed to pricing parity with OpenAI for applicable Azure OpenAI offerings.

For GPT-5.6 Luna on Standard Global (short context), the updated rate is $0.20 per 1M input tokens, $0.02 per 1M cached input tokens, $0.25 per 1M cached writes, and $1.20 per 1M output tokens.

If your deployment still displays the previous $1.00/$6.00 rates, we recommend checking the actual charges in Azure Cost Management or your invoice, rather than relying only on the public pricing page. Microsoft provides guidance on monitoring Foundry costs here:

Also, please verify the deployment type. Global Standard, Data Zone, Priority Processing, and Provisioned Throughput have different pricing models, so the rates may not be directly comparable.

If your actual billing still shows the old rates for usage after August 1, 2026, Please reach out to use by opening an Azure Support billing request for further assistance

The following references might be helpful , please check them out

Microsoft announcement — GPT-5.6 now available in Microsoft Foundry

Microsoft Foundry deployment types and pricing guidance

Plan and manage costs for Microsoft Foundry

How to create an Azure support request - Azure portal | Microsoft Learn

Thank you

Answer 1

Muhammad Waqas Khan 80

Microsoft has announced the price reduction, but the rollout is inconsistent. Some customers may see the new rates; others (especially those on Data Zone deployments) are still being billed at old rates. Check your actual invoice, not the pricing page, and open a support ticket if you're being overcharged.

The situation is fluid; monitor official Azure announcements for confirmation of when the new pricing is universally applied.

Hope this helps clarify the messy reality

0 comments

Answer 2

Hello Verma, Anand.

Greetings! Thanks for raising this question in the Q&A forum.

Good news, this has already happened. Azure OpenAI in Foundry Models has matched OpenAI's July 30, 2026 price cut for GPT-5.6 Luna and Terra. Microsoft's own Microsoft Foundry blog confirms the updated pricing table, and Azure Support has separately confirmed to other customers reporting the same question that the engineering team applied these reduced rates in Azure effective August 1, 2026.

Current GPT-5.6 Luna pricing on Standard Global (short context) is $0.20 per million input tokens, $0.02 per million cached input tokens, $0.25 per million cached writes, and $1.20 per million output tokens. This matches OpenAI's direct API pricing after the 80 percent cut.

If your deployment is still billing at the old rate ($1.00 input / $6.00 output), this is very likely a display lag rather than an actual billing issue. Several customers reported the public Azure pricing page at azure.microsoft.com/en-us/pricing/details/azure-openai showing stale numbers even though the underlying meter had already switched to the reduced rate. Check your actual invoice or Cost Management usage data rather than only the public pricing page, since that reflects what you are truly billed.

Confirm your deployment type and region, since Data Zone and Priority Processing pricing differ from Standard Global and are not fully public. Ask your Microsoft account team or Azure sales representative for exact Data Zone and Priority Processing rates for Luna if that is what you are using, as those were not published in the general announcement.

If your bill genuinely still reflects the old $1.00/$6.00 rate after August 1, 2026, open an Azure Support billing request with your subscription ID, the deployment name, region, and a recent invoice line item. Support can confirm whether re-rating is needed for your specific meter.

If this answer helps you kindly accept the answer which will help others who have similar questions.

Best Regards,

Jerald Felix.