标签：#imagen3

June 24, 2026. That is the shutdown date for every Imagen model on Firebase AI Logic — imagen-3.0-generate-002 , imagen-4.0-generate-001 , imagen-4.0-ultra-generate-001 , imagen-4.0-fast-generate-001 . All of them. If you have been putting off this migration, you have run out of runway. The replacement is Google's Gemini Image models — internally called "Nano Banana," publicly named gemini-2.5-flash-image . The migration is mostly a one-function rename and a response structure update, around 90 minutes of work for most codebases. The catch: one Imagen capability, mask-based editing, has no replacement at all. That separate deadline hits June 30. What Goes Dark and When Firebase AI Logic's migration documentation confirms these shutdown dates: imagen-3.0-generate-002 — June 24, 2026 imagen-4.0-generate-001 — June 24, 2026 imagen-4.0-ultra-generate-001 — June 24, 2026 imagen-4.0-fast-generate-001 — June 24, 2026 imagen-3.0-capability-001 (mask editing: inpainting, outpainting, object removal) — June 30, 2026 Vertex AI runs on a slightly different clock — Google recommends migrating before June 30, with a hard shutdown date of August 17 for Vertex AI users on legacy Imagen endpoints. Firebase AI Logic is the shorter deadline. Don't assume extra time if your app uses the Firebase SDK. The Core Migration: Python Three things change simultaneously: the method name, the model identifier, and the response structure. All three break if you miss any one of them. Before (Imagen): import google.generativeai as genai client = genai . Client ( api_key = " YOUR_KEY " ) response = client . models . generate_images ( model = " imagen-4.0-generate-001 " , prompt = " A red fox running through snow " , config = { " number_of_images " : 1 , " output_mime_type " : " image/jpeg " } ) image_bytes = response . generated_images [ 0 ]. image . image_bytes After (Gemini Image): import google.generativeai as genai client = genai . Client ( api_key = " YOUR_KEY " ) response = client . models . g