Access Nano Banana 2 Lite — also searched as Gemini 3.1 Flash-Lite Image or Nano Banana 2 Lite Beta — through EvoLink's unified image API. Test ~4-second text-to-image, image editing, and multi-image composition at native 1K before you integrate.
Playground
Pick a route, write a prompt, and see the estimated cost before you spend anything.
Create task
Fixed to gemini-3.1-flash-lite-imageUp to 14 images, 20 MB each. JPEG, PNG or WebP.
Generated image links stay valid for 24 hours — save anything you want to keep.
Two routes run on the same image endpoint. Official Lite bills per output image by usage; the Beta route bills a flat rate per call.
One 1K 9:16 image with one reference image
About 331 tests with $10 in credits.
The most common real request on EvoLink: the Official route, one 1K 9:16 image with one reference image. Select Official in the Playground above to try it.
Prototyping with the first-test setup above
Batch generation and prompt sweeps
High-volume pipelines and A/B sets
Estimated cost for one generated image. Final cost may vary by output quality and input images.
Official Lite bills per output image by resolution and usage. The Beta route is a flat per-call rate at 1K.
gemini-3.1-flash-lite-imageBilled per output image, plus a per-image fee for every reference image and per-1K-token fees for prompt text and thinking output. Native 1K only. Use the calculator below to price your own case.
| Item | Rule | Rate | Billing |
|---|---|---|---|
| Input Image | Applied to each reference image in image_urls, for image-to-image, editing and multi-image composition. | $0.0003 / image-15% 0.0162 cr / imageofficial price $0.0003 | Per image |
| Image generation · 1K | The generated image itself. Native 1K only — one image per call, and the fee applies to each image when you request several. | $0.029 / image-15% 1.9421 cr / imageofficial price $0.034 | Per image |
| Input Text | Produced by the prompt itself. Typically a small share of a request. | $0.0003 / 1K tokens-15% 0.0145 cr / 1K tokensofficial price $0.0003 | Per 1K tokens |
| Output Text + Thinking | Text and thinking tokens the model emits alongside the image. Scales with model_params.thinking_level. | $0.0013 / 1K tokens-15% 0.0867 cr / 1K tokensofficial price $0.0015 | Per 1K tokens |
Figures are estimates. Final charges use actual text and thinking token usage.
nano-banana-2-lite-betaOne flat rate per successful call, whatever the aspect ratio - no token accounting and nothing to estimate. Auto or aspect-ratio sizes at the 1K tier only, one image per call. Google's list price has no equivalent for this route, so no official comparison is shown.
| Item | Rule | Rate | Billing |
|---|---|---|---|
| 1K output | Any aspect ratio or auto, at the 1K tier. One image per successful call, regardless of prompt length or reference images. | $0.018 / request 1.2 cr / request | Per call |
Nano Banana 2 Lite is the community name for Google's Gemini 3.1 Flash-Lite Image model (model ID: gemini-3.1-flash-lite-image). It is the fastest, most cost-efficient tier of Nano Banana 2: image generation and editing in about four seconds at native 1K. EvoLink exposes it as a task API with two routes — Official Lite and an experimental Lite Beta — plus an in-browser playground and per-image pricing.
Both routes share the same image endpoint. Only the model parameter changes.
gemini-3.1-flash-lite-imageProduction route. Native 1K output with per-image billing that scales with resolution and usage.
nano-banana-2-lite-betaExperimental route for cost-sensitive tests. Fixed 1K output billed per call; keep Official Lite as the production fallback.
Built for fast, high-volume image generation and editing.
The fields that shape output and cost on POST /v1/images/generations. Generation is asynchronous — poll the returned task ID or pass a callback, and save results promptly since image links stay valid for 24 hours. Full schemas and code samples live in the API tab.
modelstring · required'gemini-3.1-flash-lite-image' for the Official Lite route, or 'nano-banana-2-lite-beta' for the flat-rate Beta route.
promptstring · requiredThe text instruction describing the image to generate, or how to edit the input images. Prompt tokens are metered as text input on the Official Lite route.
sizestring · default auto'auto' or one of 14 aspect ratios — 1:1, 1:4, 4:1, 1:8, 8:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.
qualitystring · 1K onlyOutput resolution. Nano Banana 2 Lite is native 1K only — 2K and 4K are not available on this tier.
image_urlsarray · up to 14Reference image URLs for image-to-image, editing, and multi-image composition. Max 20 MB per image; .jpeg, .jpg, .png or .webp; URLs must be reachable by the server. Each one adds an input-image line item on Official Lite.
model_params.thinking_levelstring · auto / min / highNested in model_params. Controls how much reasoning the model does before generating — min is fastest, high favours quality. Official Lite route only.
callback_urlstring · optionalHTTPS webhook that receives the task result on completion, failure, or cancellation, as an alternative to polling. Sent after billing is confirmed; max 2048 chars, 10s timeout, up to 3 retries. Internal IPs are rejected.
Use the EvoLink API for product integration and batch jobs, or call it from Codex, Claude, or Gemini for fast creative and development workflows. Both paths share the same EvoLink API key, balance, model routes, and task history.
Best for: product backends, batch jobs, automated workflows
Call EvoLink’s unified image API from your server and control the model ID, parameters, task queue, callbacks, and result storage.
Best for: creative and development tasks in Codex, Claude, and Gemini
Give the Agent your output goal, assets, and acceptance criteria. It can choose the route, assemble the request, track the task, and return the result without requiring you to hand-code every step.
Four-second round-trips keep chat-driven editors and prompt-remix UIs responsive.
The lowest per-image cost in the family makes large catalogue and variant sweeps affordable.
Swap colours, build stickers, and adjust backgrounds inline — no masks or layer management.
14 aspect ratios cover feed, story, portrait, and cinematic crops from one model.
This example shows the shortest runnable flow: create a task, save the returned task ID, then poll or wait for an HTTPS callback. Open the API tab for the complete parameter and response reference.
curl -X POST https://api.evolink.ai/v1/images/generations \
-H "Authorization: Bearer $EVOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.1-flash-lite-image",
"prompt": "A neon-lit street market at night, cinematic",
"size": "16:9",
"quality": "1K",
"callback_url": "https://your-domain.com/webhooks/image-task-completed"
}'
# Save the returned task ID, then query:
curl https://api.evolink.ai/v1/tasks/{task_id} \
-H "Authorization: Bearer $EVOLINK_API_KEY"Check every value against the parameter cards above: ranges, allowed options, asset counts, and file types must match the selected route.
Check the Authorization bearer token and confirm the available balance and task billing in the console.
Review likeness rights, brand assets, readable text, sensitive content, and input-asset compliance.
A Nano Banana 2 Lite task that reaches a final failed state is not charged. Keep the task ID, inspect the final state, and identify whether the cause is validation, review, or execution before retrying.
callback_url must be a public HTTPS URL, not localhost or a private network address. Always keep the task ID as a polling fallback.
When throughput and cost matter more than maximum resolution, Lite is the tier that keeps iteration fast and per-image spend low.
| Dimension | Nano Banana 2 Lite | Nano Banana 2 |
|---|---|---|
| Engine | Gemini 3.1 Flash-Lite Image | Gemini 3.1 Flash Image |
| Speed | ~4s per image (≈2.7× faster) | Standard (baseline) |
| Resolution | 1K (1024×1024) only | 1K / 2K / 4K |
| Editing and multi-image | Yes | Yes |
| Best for | High-volume, rapid iteration, lowest cost | Maximum quality, 2K/4K production finals |
Official Lite (gemini-3.1-flash-lite-image) and Lite Beta (nano-banana-2-lite-beta) share POST /v1/images/generations. Only the model parameter differs — no separate base URL or key.
The create call returns a task ID; poll GET /v1/tasks/{task_id} until it reaches a terminal state, or pass callback_url to be notified on completion, failure, or cancellation. Callbacks to internal IPs are rejected.
Official Lite bills per output image by usage, plus input images and input/output text tokens as separate line items. Lite Beta is a flat rate per successful call at 1K. Failed tasks are not charged.
Generating at volume gets you more iterations for the same budget and wall-clock time. When you do not need 2K or 4K, Lite keeps latency and cost down without changing your prompts or pipeline.
Character consistency holds across edits, so people, mascots, and products stay recognizable from first draft to final — with no masks or layers to babysit.
Reach Nano Banana 2 Lite through EvoLink with a single API key. No Google Cloud project, no billing configuration, no quota requests — and the same OpenAI-compatible request format you already use.
Other Nano Banana image models available on EvoLink.
Nano Banana
The original Gemini 2.5 Flash Image model: fast generation and editing, multi-image blending, and consistent characters.
View APICompare throughput, resolution, and per-image cost across the catalogue.
Grok Imagine Image 2.0
xAI image model: text-to-image and image editing on one route, with 1K / 2K output.
View APIModel identity, resolution limits, speed, and billing.
Nano Banana 2 Lite is the community name for Google's Gemini 3.1 Flash-Lite Image model (model ID: gemini-3.1-flash-lite-image). It is the fastest, most cost-efficient variant of Nano Banana 2, generating images in about 4 seconds at native 1K output.
Nano Banana 2 supports higher-resolution 1K, 2K, and 4K output for maximum quality. Nano Banana 2 Lite is optimized for speed and cost: about 4-second generation (roughly 2.7× faster) at native 1K only (no 2K/4K). Choose Lite for high-volume, latency-sensitive work, and the standard model when you need higher resolution.
Output is native 1K (1024×1024) only — 2K and 4K are not supported. It supports 14 aspect ratios, including 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, and 21:9.
Nano Banana 2 Lite generates a text-to-image result in about 4 seconds — roughly 2.7× faster than Nano Banana 2 — making it suitable for interactive editors, chat-driven tools, and high-throughput batch pipelines.
Google's published rate is about $0.034 per 1K image. EvoLink offers two routes: the Official Lite route bills per output image, plus a small per-image fee for each reference image and per-1K-token fees for prompt text and thinking; the Beta route is one flat rate per call at 1K. The pricing section shows both.
Yes. Every image carries an always-on SynthID watermark for provenance.
Sign up on EvoLink, get an API key, and call the Nano Banana 2 Lite endpoint. No Google Cloud account needed. EvoLink provides OpenAI-compatible endpoints, so integration takes minutes.