What’s new
- The speed tier of the Imagen 4 family: launched 2025-08-15 alongside the general availability of the whole Imagen 4 family, officially "built for speed" and aimed at rapid, high-volume generation.
- Resolution and text rendering both improved over Imagen 3: supports up to 2K resolution, with meaningfully better text rendering per Google.
- Google now frames it as the specialist option: Vertex AI's own docs recommend starting with Gemini (the Nano Banana family) for image generation by default, and reaching for Imagen 4 when you specifically need its image quality, texture, or branding/logo/layout strengths.
- Covers the common social/ad/TV aspect ratios: 1:1, 16:9, 9:16, 3:4 and 4:3, defaulting to 16:9.
What it’s good for
- Generating a lot of candidate images quickly.
- Work that cares about image quality but prioritizes speed over the absolute best detail.
- Static, well-composed brand assets — logos and layout work, which Google specifically calls out as an Imagen 4 strength.
Prompting tips
- Structure prompts in three parts — subject, context, style: say what's in the shot, where it's placed, then what artistic approach you want.
- Use photography language directly: shot size (close-up/wide), lighting (natural/dramatic/warm/cold), lens type (macro/fisheye/wide-angle/35mm), and "bokeh" all work as direct prompt terms.
- Add quality modifiers: terms like "high-quality," "4K," or "professional photography" help push the result further.
- State exclusions plainly: write "exclude: walls, frame" rather than "no walls" or "don't include a frame."
- Keep on-image text to 25 characters or fewer: longer text tends to render poorly; you can also note the general font style and size.
- Name the art style directly: phrases like "a painting of..." or "a sketch of...", or naming a specific movement (impressionist, renaissance, pop art), give more consistent results.
A product illustration of a sneaker, high-quality, 4K, professional studio lighting, warm natural
light, 35mm lens, shallow depth of field, bokeh background. 4:3. Exclude: people, text, frame.
Known limits
- Pure text-to-image — no editing support: there's no reference/input image field in the API, so it can't edit or work from an existing image.
- Complex prompts combined with automatic prompt enhancement can misfire: Google's own docs note that Imagen 4 Fast may produce undesirable results on complex prompts when prompt enhancement is on, and recommend turning it off in that case.
- On-image text is recommended to stay under 25 characters: Google's guide suggests this limit — longer text tends to render poorly.
Using it in Nomi
On the generation canvas, add a image card, pick “Imagen 4 Fast” in the model picker and choose a mode (Text to image). Add your references, write the prompt and generate. In the storyboard you can pick it per shot. The first time, connect any one of the providers above under Settings → Models.
Sources
- Announcing Imagen 4 Fast and the general availability of the Imagen 4 family (Google Developers Blog, 2025-08-15)
- Imagen 4 overview (Google Cloud Vertex AI official docs)
- Generate images with Imagen (Google Cloud Vertex AI official docs)
- Imagen prompt and image attribute guide (Google Cloud Vertex AI official docs)
- Imagen 4 Fast API reference (Kie.ai)
