nano-banana-2-newFast AI image generation, semantic editing, and Google Search grounded visual workflows for scalable production.
1. Get your API key: Create an API key in your dashboard and start integrating immediately. Get API Key
2. Add credits: Fund your account before running image jobs. Add Credits
3. Choose your workflow: Use nano-banana-2-new for text-to-image generation, or nano-banana-2-new-edit for semantic editing and composition. Pass prompt, resolution, size, and optional google_search.
4. Submit and poll asynchronously: In edit mode, send at least one image_urls entry and up to 14 references. The current route uses an async workflow with polling and keeps n fixed at 1, with no local mask workflow exposed. See the docs for request examples: Nano Banana 2 API Docs
Instant Access to Nano Banana 2 API with no waitlist required. The Ultimate Engine for High-Speed Visual Intelligence combines flash-speed inference, semantic image editing, Google Search grounding, and multi-image composition for ultra-low-cost, enterprise-scale image production.
A high-speed image engine for generation, editing, grounding, and scalable composition.
Nano Banana 2 is designed for low-latency creative workflows, helping teams move from prompt to usable asset with flash-speed inference and high-fidelity visual synthesis.
Google Search grounding helps Nano Banana 2 generate more context-aware visuals from real-world references, improving relevance for ads, products, and localized content.
Perform localized visual changes through natural-language instructions alone. Nano Banana 2 supports zero-mask semantic editing, so teams can restyle, replace, or refine regions without manual mask drawing.
Combine multiple reference images in one workflow to preserve subject identity, style direction, and scene coherence. This makes Nano Banana 2 especially strong for high-volume composition tasks.
Generate product scenes, ad variants, and catalog visuals at scale with low-latency production and consistent output quality.
Create localized campaign visuals with grounded references, text-aware generation, and fast semantic edits for different markets.
Build editing tools that accept natural-language instructions for restyling, retouching, and scene-level visual adjustments.
Merge references into a single branded output while preserving style direction, subject details, and composition consistency.
Power user-facing creative apps with fast image generation, style transfer, and prompt-driven editing for daily production work.
Run large-scale automated image pipelines for merchandising, content personalization, and creative operations.
Compare Nano Banana 2 with other popular image generation models available on Vidgo.
| Feature | Nano Banana 2 | Nano Banana Pro | Nano Banana | GPT Image 1.5 |
|---|---|---|---|---|
| Powered By | Gemini 3.1 Flash | Gemini 3 Pro | Gemini 2.5 Flash | GPT-4o |
| Best For | Fast bulk generation (default 2K) | Complex reasoning + quality | Quick iteration + edits | Speed + precise edits |
| Resolution | 1K / 2K (default) / 4K | Up to 4K (default 1K) | Standard | High-res |
| Text Rendering | Excellent (multi-language) | Strong | Good | Strong |
| Reference Images | Up to 14 | Up to 14 | Up to 5 | 1 |
| Aspect Ratios | Standard + extreme (1:8, 8:1) | Standard | Standard | Standard |
| Image Editing | Yes (nano-banana-2-new-edit) | Yes (nano-banana-2-edit) | Yes | Yes |
| Price on Vidgo | 5 credits ($0.025) | 10 credits ($0.05) | 5 credits ($0.025) | 2 credits ($0.01) |
1. Get your API key: Create an API key in your dashboard and start integrating immediately. Get API Key
2. Add credits: Fund your account before running image jobs. Add Credits
3. Choose your workflow: Use nano-banana-2-new for text-to-image generation, or nano-banana-2-new-edit for semantic editing and composition. Pass prompt, resolution, size, and optional google_search.
4. Submit and poll asynchronously: In edit mode, send at least one image_urls entry and up to 14 references. The current route uses an async workflow with polling and keeps n fixed at 1, with no local mask workflow exposed. See the docs for request examples: Nano Banana 2 API Docs
Start with $0.025/1K image for 5 credits, scale to $0.040/2K image for 8 credits, or run $0.060/4K image for 12 credits.
Built for high-volume automated workflows and enterprise-scale production. Nano Banana 2 supports 10 aspect ratios, optional Google Search grounding, and up to 14 reference images in the edit workflow.
The API uses an async generation workflow with predictable credit billing, no subscription requirement, and cost-efficient scaling for fast image generation and semantic editing.
Start immediately with a single API key and no waitlist, so teams can move from testing to production faster.
Tiered 1K, 2K, and 4K pricing keeps costs efficient for automated image generation and large-volume editing workloads.
Optional Google Search grounding improves contextual relevance for real-world, market-aware, and localized visual content.
Natural-language editing lowers operator friction and speeds up creative iteration without requiring manual mask authoring.
Async processing, scalable billing, and predictable request patterns support high-volume visual automation pipelines.
Use one platform and one billing flow to access Nano Banana 2 alongside the rest of your image and multimodal model stack.
<p>Nano Banana 2 uses the <strong>nano-banana-2-new-edit</strong> workflow to combine text instructions with reference images. This allows teams to restyle scenes, refine products, or adjust composition without rebuilding the image from scratch.</p>
<p>Zero-Mask Semantic Editing means you can describe the change in natural language and let the model handle localized edits without manually painting a mask. It is designed for faster, lower-friction editing workflows.</p>
<p>You can use up to <strong>14 reference images</strong> to ensure maximum character and style consistency.</p>
<p>Yes. Its low-latency generation, scalable pricing tiers, and semantic editing workflow make it well-suited for catalog refreshes, ad variants, and merchandising automation.</p>
<p>Yes. The new model family supports optional <strong>Google Search grounding</strong>, which helps create more context-aware visuals based on web-grounded references.</p>
<p>The current route supports <strong>1K</strong>, <strong>2K</strong>, and <strong>4K</strong> output, along with <strong>10 aspect ratios</strong> including 1:1, 16:9, 9:16, 21:9, and more.</p>
<p>Pricing starts at <strong>5 credits</strong> for <strong>1K</strong> at <strong>$0.025</strong>, rises to <strong>8 credits</strong> for <strong>2K</strong> at <strong>$0.040</strong>, and <strong>12 credits</strong> for <strong>4K</strong> at <strong>$0.060</strong>.</p>
<p><strong>nano-banana-2-new</strong> is optimized for fast text-to-image generation with optional Google Search grounding, while <strong>nano-banana-2-new-edit</strong> is built for semantic editing and multi-image composition with up to 14 references.</p>
<p>Yes. Nano Banana 2 is built for high-throughput API usage with predictable credit billing, async processing, and workflow flexibility for automated production systems.</p>
