Nano Banana Pro availability should be checked by access route, not by a single universal platform list. The model name is gemini-3-pro-image, but the controls, billing, quota, data terms, and regional behavior depend on where you call it.
The safest current answer is:
| Route | Use it when | Trust boundary |
|---|---|---|
| Gemini API through Google AI Studio | You are prototyping, testing prompts, or building with a developer API key. | Check Google's Gemini API docs, pricing page, and the active AI Studio project. |
| Vertex AI / Gemini Enterprise Agent Platform | You need enterprise deployment, Google Cloud controls, data residency, security, or provisioned capacity. | Check Google Cloud model pages, locations, quotas, and billing. |
| Gemini app and Google product surfaces | You want a consumer or creator workflow without writing code. | Check the product UI, plan, region, and account eligibility at the time of use. |
| Workspace or embedded Google product routes | You need images inside slides, docs, video, research, or enterprise productivity flows. | Check the specific product admin console and release notes. |
| Third-party gateway | You need payment, SDK, or multi-model routing convenience. | Verify current model coverage, billing, logs, and support terms directly in that provider. |
Do not rely on a static "available on 12 platforms" table. Platform surfaces change faster than a blog page, and the wrong table can create false expectations for Search Console, users, and support teams.
What is currently safe to say
Checked on July 8, 2026, these facts are stable enough to use in a route guide:
| Fact | Source boundary |
|---|---|
Nano Banana Pro maps to Gemini 3 Pro Image, model ID gemini-3-pro-image. | Google's Gemini image generation docs and pricing pages. |
| Gemini API image generation examples now use the Interactions API for current Gemini 3 image models. | Google's image generation docs. |
| Gemini API pricing and free/paid status are model-row specific. | Google's Gemini API pricing page. |
| Gemini API rate limits are project-owned, not API-key-owned. | Google's rate-limit docs. |
| Vertex AI / Gemini Enterprise model pages can expose cloud-specific launch stage, locations, and security controls. | Google Cloud model documentation. |
Everything else should be treated as a live product check: exact consumer app quotas, country availability, Workspace plan entitlements, app menu labels, promotional trials, and third-party pricing.
Route 1: Google AI Studio and Gemini API
This is the cleanest developer starting point. Use it when you want to test prompts, inspect current model availability, and turn a working prompt into code.
Use this route when:
- you need a simple API-key based developer workflow;
- you want to compare
gemini-3.1-flash-imageandgemini-3-pro-image; - you need current pricing and rate-limit context before building;
- you are not yet ready for a full Google Cloud enterprise deployment.
Do not assume AI Studio gives the same quota, policy, or cost boundary as the Gemini app. A consumer subscription does not automatically define API access. The API project, billing state, and model row define the developer contract.
Before you call the model, verify:
| Check | Why it matters |
|---|---|
| Does the pricing page show the model row you need? | Some models and surfaces may be paid-only. |
| Does the active AI Studio project show available quota? | Limits belong to the project, not the key. |
| Are you using the current model ID? | Older preview names should not be copied into new code. |
| Are you using current API shape? | Current Gemini 3 image examples use Interactions API patterns. |
| Do you need 1K, 2K, or 4K output? | Resolution changes cost, latency, and review expectations. |
For model and cost details, use the current Nano Banana vs Nano Banana Pro comparison and the Nano Banana Pro pricing quota guide.
Route 2: Vertex AI and Gemini Enterprise Agent Platform
Use the Cloud route when the question is not just "can I generate an image?" but "can my organization run this in production under Google Cloud controls?"
This route fits:
| Need | Why Cloud route fits |
|---|---|
| enterprise deployment | Cloud projects, IAM, audit, and organization policy matter. |
| data residency or security controls | Google Cloud model pages expose supported controls and locations. |
| quota planning | Paid project quotas, model-specific limits, and increase requests are managed in Cloud context. |
| provisioned capacity | High-volume production may need capacity planning instead of best-effort interactive access. |
| compliance review | Enterprise teams need docs, regions, logging, and procurement records. |
Google Cloud documentation lists Gemini 3 Pro Image as a generally available model in the Gemini Enterprise Agent Platform model surface, with model availability and security-control details handled by the Cloud docs. Treat those Cloud docs as the authority for launch stage, supported regions, data residency, CMEK, VPC-SC, audit, provisioned throughput, and retirement timing.
The Cloud route can be the correct answer even when AI Studio works. AI Studio is excellent for exploration. Vertex AI or Gemini Enterprise is the production route when governance, support, deployment controls, and procurement are the reader's real job.
Route 3: Gemini app and consumer product surfaces
Consumer surfaces can be useful for creators, marketers, and quick visual exploration, but they should not be described like API surfaces.
Use consumer/product UI only when:
- a human is creating images manually;
- the output will be reviewed before publication;
- app-level plan, region, and account limits are acceptable;
- no application depends on programmatic access.
Avoid these claims unless you verified them in the product UI on the same day:
| Avoid claiming | Safer wording |
|---|---|
| "available globally" | "check the Gemini app or product UI for your account and region." |
| exact daily image counts | "quota depends on the account, plan, and current product policy." |
| exact subscription features | "plan-specific features can change; verify before purchase." |
| "same as API" | "consumer app access and API access have different contracts." |
For SEO and GEO, this distinction matters. A user asking "where can I use Nano Banana Pro?" may need either a human product entry point or a developer API entry point. Mixing them into one table makes the answer look comprehensive while making it less reliable.
Route 4: Workspace and embedded Google product routes
Workspace and embedded product routes should be treated as task surfaces, not as the source of model truth. A presentation, video, research, or design product may expose Nano Banana Pro-powered generation, but the user's available controls depend on admin policy, plan, release track, region, and product state.
Use this route when:
| User job | Better surface |
|---|---|
| presentation visuals inside an existing deck | the presentation product where the deck lives |
| internal marketing drafts | Workspace or Gemini app, if enabled |
| enterprise document workflow | Workspace admin-approved surface |
| automated asset pipeline | Gemini API or Vertex AI, not a manual product UI |
If you cannot confirm a product-specific entitlement from current docs or UI, do not publish a hard availability claim. Say plainly that the user should check the product menu, admin console, or release note for their account.
Route 5: third-party gateways
A gateway route can help when the reader needs an OpenAI-compatible API shape, local payment flow, multi-model routing, or a simpler operations path. It should not be framed as an official availability source.
Before recommending a gateway, verify:
- current support for
gemini-3-pro-image; - whether the provider supports the output sizes the reader needs;
- current pricing, minimum top-up, and billing logs;
- whether failed or blocked generations are charged;
- latency, retry behavior, and 429 semantics;
- support response and refund policy;
- whether the route is appropriate for the user's data sensitivity.
Use the official Google route when model truth, compliance, quota ownership, or enterprise support is the primary concern. Use a gateway only when its operating tradeoff materially helps the reader.
For a route-specific gateway decision, use the current Nano Banana Pro API route guide and verify the provider dashboard in the same session.
Availability checklist
Before telling a user "yes, Nano Banana Pro is available," answer these six questions:
| Question | Why it blocks wrong recommendations |
|---|---|
| Which route do they need: app, API, Cloud, Workspace, or gateway? | Availability is route-specific. |
| What exact model ID is required? | Nano Banana naming has changed across releases. |
| Is the account in a supported project, plan, or region? | UI availability can differ by account. |
| Does the route support the required output size? | 1K, 2K, and 4K have different costs and controls. |
| Are rate limits acceptable for the workload? | Project quota can block an otherwise available model. |
| Does the route fit the data and compliance boundary? | Consumer app, API, Cloud, and gateway terms differ. |
If any answer is unknown, qualify the recommendation instead of writing a confident availability claim.
What changed from older guides
Older Nano Banana Pro availability pages often overreached in four ways:
| Old pattern | Current fix |
|---|---|
| long platform matrix with exact feature claims | route map with explicit verification owner |
| consumer app quotas treated as API facts | app, API, and Cloud contracts separated |
| preview model IDs in code examples | current stable model ID and current docs linked |
| gateway price claims with affiliate links | gateway kept as a verified operating route, not official truth |
This page intentionally does less platform enumeration than a generic roundup. That is the point: a shorter route map with live verification boundaries is safer for users and less likely to create duplicate, stale, or over-optimized search pages.
FAQ
Is Nano Banana Pro available in Google AI Studio?
Use AI Studio and the Gemini API docs to check the current model row and your active project. For developer testing, this is usually the first place to verify API access before writing production code.
Is Nano Banana Pro available on Vertex AI?
Yes, use the current Google Cloud model documentation for the enterprise route. Cloud docs are the authority for model launch stage, supported locations, security controls, quotas, and production deployment details.
Does Gemini app access mean API access?
No. A consumer app or subscription route is not the same contract as Gemini API or Vertex AI. API work needs its own project, billing, quota, and model availability checks.
Should I trust a platform matrix from an older article?
No. Use it only as a clue. Verify the exact route, account, region, plan, model ID, output size, and billing state before relying on it.
Which route should a developer choose first?
Start in Google AI Studio for prompt and API validation. Move to Vertex AI or Gemini Enterprise when production governance, security controls, quota planning, or procurement matters.
Which route should a non-technical creator choose first?
Start with the Gemini app or the Google product surface where the work already lives. Verify the visible model options and plan limits in the product UI before assuming Pro access.
Bottom line
Nano Banana Pro availability is not a single yes/no answer. It is a route decision.
Use Gemini API and AI Studio for developer validation, Vertex AI or Gemini Enterprise for governed production, Google app/product surfaces for manual creation, and gateways only when their current dashboard and support terms solve a real operating problem. Keep official Google docs as the source of model truth, and qualify anything that depends on account, plan, region, or provider state.
