Failed to deploy model gpt-6-luna

ak-vvp 25 Reputation points
2026-11-02T10:47:54.6033333+00:00

I was requested to deploy new additional gpt-6-luna model for which I received an error copied below.

Failed to deploy model gpt-6-luna
715-123420: Our system has detected this request as unusual activity for your account. If you are confident this is in error, please contact support.

This model was supposed to expand the existing list of other models I have deployed previously, which we are already using with our internal tools. None of the earlier models returned an error, and a newer gpt-6.1-sol was deployed successfully as well.

Below are the error IDs I received. I would appreciate any help with this deployment, as this seems to be the only support path I can use. Thank you.

Trace ID : 35a7593b-3236-4be0-85df-419c0b4dcf95
Client request ID : 267dc3ba-58c0-465f-aa5f-5d18177b4bb9
Service request ID : 642a4498-0c70-424b-a888-455414024b01

Foundry Models
Foundry Models

A catalog of AI models in Microsoft Foundry that you can discover, compare, and deploy using Azure’s built‑in tools for evaluation, fine‑tuning, and inference

0 comments
Answer accepted by question author
Hemant Bharadwaj 355 Reputation points Microsoft Employee
2026-11-02T16:00:27.6933333+00:00

Likely cause/summary: error 715-123420 is typically a service-side risk/abuse protection decision for that specific deployment request, not a normal quota-exceeded error. Since other models in the same subscription worked, this is likely request/context-specific.

  1. Verify this is not a quota/capacity configuration issue first.
  2. Retry with controlled variations (one change at a time).
    • Use a different deployment name.
    • Try the same model in another supported region.
    • Try from the same subscription/resource with no other changes.
    • Record each attempt timestamp in UTC.
  3. Keep evidence for escalation (you already have key IDs).
    • Trace ID: 35a7593b-3236-4be0-85df-419c0b4dcf95
    • Client request ID: 267dc3ba-58c0-465f-aa5f-5d18177b4bb9
    • Service request ID: 642a4498-0c70-424b-a888-455414024b01
    • Include subscription ID, resource name, region, model/version, and deployment type.
  4. Open Azure support for backend review/unblock.

Confirmed facts from your post:

  • gpt-6-luna deployment failed with 715-123420.
  • Other model deployments in your environment succeeded (including gpt-6.1-sol).

Possible causes (not yet confirmed):

  • policy/risk trigger on this exact deployment attempt,
  • temporary backend gating for model/region/subscription combination,
  • model-specific capacity or eligibility checks not reflected as standard quota errors.

Validation checks:

  • deployment succeeds with one controlled variation (name/region/type), or
  • support confirms root cause and clears the block.

“If you find this answer helpful, please upvote it. If this resolves your issue, please mark it as the accepted answer so that it can help others in the community.”

Was this answer helpful?

1 person found this answer helpful.
0 comments

1 additional answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. ak-vvp 25 Reputation points
    2026-11-03T16:56:08.5166667+00:00

    Hi Hemant, thank you for your reply.

    I was going through your list of suggestions. I still have available quota for the gpt-6-luna model deployment, and the model is available in my selected region with given type. I have not found any other obvious limits.

    It took several attempts to create the model with different parameters, every attempt failed until I was successful with adding the same model with a different name and different type. On that basis I tried to create the model as I originally needed and it was successful as well.

    I don't know what was the reason for success, or if the remote-side block was only temporary. In the end I achieved what I wanted and this topic can be closed.

    Thank you again for your help.

    Was this answer helpful?

    1 person found this answer helpful.

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.