Google's update to Nano Banana 2: better images, cleaner text and steadier edits at the same Flash speed. Paste a prompt list and get the whole set back as one ZIP.
One per line — each line is a separate prompt.
Drag & drop or click to upload — PNG / JPEG / WEBP / GIF, up to 10MB each.
A point release on Nano Banana 2, the model most people already use through Gemini. Everything below comes from Google's own Gemini API documentation (October 2026), not from us.
Google calls 2.1 an update to Nano Banana 2 that keeps Flash-level speed and cost, and positions it as the everyday workhorse for creating and editing images. Nano Banana Pro stays the premium option for the hardest visual work, so the family now splits neatly: 2.1 for volume, Pro for the few images that need the most care.
The list of improvements is short and specific: visual quality, text rendering, consistency across editing turns, and Google Search grounding, which now draws on Google Image Search as well as the web. Readable text and a look that survives several rounds of edits are exactly where batches usually break, so these are the right things to fix for anyone generating sets.
Per Google, the model renders at 1K, 2K and 4K and accepts up to 10 object references, 4 character references and 3 style references. A provider may expose fewer slots than the model supports, so the limits that apply here are the ones in the table below, read from our live configuration.
Ratios, resolutions, reference limits and credit prices are read from our live model configuration, so the table always matches what the form accepts.
Everything in this table, the price included, comes from the running configuration, not from this page's copy.
Three steps, the same as for Nano Banana 2, Pro and every other model on BulkImagen. The choices that are specific to this one are in the middle step.
Paste one prompt per line or import a CSV. Only have an idea? Write it as one sentence and press Write with AI under the prompt box to turn it into one prompt per image.
Attach product, character or style references if the set has to stay consistent, then pick the aspect ratio and resolution. The image count and credits show up before anything is generated.
All prompts render in parallel. Retry any image that misses, then download the whole batch as one ZIP, each file named after its prompt.
Where the changes Google lists should matter most when you generate in bulk.
Posters, social cards, labels and thumbnails, where one misspelled word means regenerating the set. Text rendering is one of the headline improvements in this release, and it pays off most when every image in a batch carries copy.
One product or one character across dozens of scenes. With several object and character reference slots, the model is built for sets where the subject must stay identical while the scene changes.
Ad variations, catalog backgrounds, illustration series. Flash speed and cost make 2.1 the one to run a long list on first, keeping Nano Banana Pro for the handful of images that need extra care.
Anything else? Email support@bulkimagen.com
Google's newest Flash image model, a whole prompt list at a time: paste the lines, check the credits, take the ZIP.