The notice
On June 2, 2026, OpenAI announced retirement of gpt-image-1-mini, gpt-image-1.5, and chatgpt-image-latest, with shutdown scheduled for December 1, 2026.
Group dependencies by their quality needs
A low-cost draft generator and a detailed product editor should not automatically receive the same replacement. Inventory what each workflow needs, the volume it handles, and the amount of human review it already receives.
Assign a supported candidate to each category. Check both access and request compatibility before making a commercial forecast based on the new selection.
Give budget owners a useful comparison
Present total candidate spend and approved assets for the same briefs. Explain any change in output settings, review standards, or retry behavior. That makes it possible to separate the effect of the model from changes to the workflow.
Do not promise that a replacement preserves all historical costs or outputs. Publish a new estimate based on the tested request pattern and the applicable rate.
Set a date before the provider deadline
Leave time for integration checks, staged rollout, and documentation updates. Track actual request identifiers after deployment so that an overlooked background job does not surface only on shutdown day. As of publication, this deadline is upcoming; revisit the official schedule before executing the migration.
Official sources
Provider announcements describe the underlying APIs. Check the live XMH.NET catalog for the models and request formats available to your account.