Nano Banana Image Generation and Editing

Nano Banana is Inference Space's integration of the Gemini image series. It includes three models — Nano Banana 2 (Small Banana, fast), Nano Banana Pro (Big Banana, quality-first), and Nano Banana 2 Lite (Small Banana Lite, ultra-fast and low-cost) — supporting text-to-image generation and reference-image editing (image-to-image), with up to 14 reference images per edit.

Parameters and capabilities align with Google's official Gemini image documentation: Gemini 3.1 Flash Lite Image and Image generation · Aspect ratios and image sizes.

The gateway provides two equivalent protocols, with the native Google protocol as the primary interface:

• Native Google (recommended): POST /v1beta/models/{model}:generateContent, directly compatible with the google-genai SDK and Gemini-compatible tools.
• OpenAI-compatible: POST /v1/images/generations and /v1/images/edits, allowing existing OpenAI image clients to connect without code changes.
• Async jobs: POST /api/jobs; use async submission and polling for long-running jobs (4K / multi-image fusion). It shares one protocol with the other image models; see Image generation and editing · Async jobs.

All three interfaces use the same authentication (Authorization: Bearer $INFERENCE_SPACE_API_KEY) and billing; choose whichever fits your existing SDK best.

Choosing a model

All three models share every interface; only the model field differs. Choose according to the scenario:

There is also the previous-generation gemini-2.5-flash-image (alias nano-banana), which supports only 1K; any other tier returns 400. New integrations should use the three models in the table above.

Native Google protocol (generateContent)

The native Google Generative Language (Gemini) protocol is the preferred interface for this series. Point the Base URL of the google-genai SDK or any Gemini-compatible tool to https://cn.inf.space.

• Text-to-image / image editing: POST https://cn.inf.space/v1beta/models/{model}:generateContent
• Streaming form: POST https://cn.inf.space/v1beta/models/{model}:streamGenerateContent?alt=sse (for tools that can only call the streaming endpoint; the image is delivered in one piece once generation finishes)
• Model list: GET https://cn.inf.space/v1beta/models

Text-to-image

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/v1beta/models/gemini-3.1-flash-image:generateContent" \
  -H "Authorization: Bearer $INFERENCE_SPACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{ "parts": [{ "text": "A seaside café terrace at sunset, warm tones, cinematic landscape composition" }] }],
    "generationConfig": {
      "responseModalities": ["IMAGE"],
      "imageConfig": { "aspectRatio": "16:9", "imageSize": "2K" }
    }
  }'

The image is in candidates[0].content.parts[].inlineData.data (raw base64; when the organization has enabled "Force image responses to use URLs" for Gemini, it is a fileData.fileUri link instead):

{
  "candidates": [
    { "content": { "role": "model", "parts": [
      { "inlineData": { "mimeType": "image/png", "data": "iVBORw0KGgo..." } }
    ] } }
  ],
  "usageMetadata": { "totalTokenCount": 1290 }
}

For Big Banana, replace gemini-3.1-flash-image with gemini-3-pro-image; everything else is unchanged. For Small Banana Lite, replace it with gemini-3.1-flash-lite-image, and imageSize can only be 1K (higher tiers are forced back down).

Image-to-image and image editing

Place the reference image in an inlineData part inside contents[].parts and submit it together with the text. For multi-image fusion, include multiple inlineData parts (refer to them in order in the prompt as "image 1 / image 2"; up to 14 images, png / jpeg / webp). Images already on the public internet can also be passed as links with fileData: { "mimeType": "image/png", "fileUri": "https://..." } (each ≤ 35MB; gs:// links are not supported):

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/v1beta/models/gemini-3-pro-image:generateContent" \
  -H "Authorization: Bearer $INFERENCE_SPACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{ "parts": [
      { "inlineData": { "mimeType": "image/png", "data": "'"$(base64 -i banana.png)"'" } },
      { "text": "Add a small straw party hat to the top of the banana and keep everything else unchanged" }
    ] }],
    "generationConfig": { "responseModalities": ["IMAGE"], "imageConfig": { "aspectRatio": "1:1", "imageSize": "1K" } }
  }'

Sizes and aspect ratios

The Gemini series supports only discrete aspect ratios and quality tiers; it does not provide an arbitrary widthxheight pixel contract. First choose an available aspect ratio for the model, then choose 1K / 2K / 4K. Supported aspect ratios vary by model (data aligned with the official Google documentation):

• The native ratio field is imageConfig.aspectRatio; the OpenAI-compatible interface uses the extension field aspectRatio. The default is 1:1.
• The native quality field is imageConfig.imageSize; the OpenAI-compatible interface uses size: 1K / 2K / 4K. Small Banana Lite outputs 1K only; the gateway forces higher tiers back to 1K.
• Only Small Banana supports the ultra-wide / ultra-tall banner ratios (1:4, 4:1, 1:8, 8:1). Big Banana and Small Banana Lite do not support these four ratios; an unsupported value returns a parameter error or uses the nearest ratio.

The following are reference pixel dimensions for each tier while preserving the aspect ratio (aligned with the official documentation; actual output governs):

• The table shows reference pixels. Different models may return slightly different pixels while preserving the aspect ratio; a 1K landscape image is not guaranteed to be exactly 1024 pixels wide.

Async jobs (/api/jobs)

For long-running generation such as 4K and multi-image fusion, use async jobs. A native Gemini request body can be submitted as-is; just add model at the top level (the async endpoint URL has no model segment):

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/api/jobs" \
  -H "Authorization: Bearer $INFERENCE_SPACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3-pro-image",
    "contents": [{ "parts": [{ "text": "A cyberpunk city at night, 4K cinematic quality" }] }],
    "generationConfig": { "imageConfig": { "aspectRatio": "16:9", "imageSize": "4K" } }
  }'

Once you have the job id, poll with GET /api/jobs/{id}; on completion, outputs[].url holds the download links. For job statuses, cancellation, listing, and the 7-day retention period, see Image generation and editing · Async jobs.

OpenAI-compatible protocol (/v1/images/*)

Existing OpenAI image clients can connect without code changes by switching only the Base, Key, and model.

• Text-to-image: POST https://cn.inf.space/v1/images/generations, application/json.
• Image-to-image / editing: POST https://cn.inf.space/v1/images/edits, multipart/form-data (image / repeated image[]), or JSON with images (public URL / data URL / base64), up to 14 images.

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/v1/images/generations" \
  -H "Authorization: Bearer $INFERENCE_SPACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gemini-3.1-flash-image", "prompt": "A seaside café terrace, cinematic landscape", "size": "2K", "aspectRatio": "16:9" }'

The response uses the standard OpenAI shape, with the image in data[0].url by default; for inline base64, send "response_format": "b64_json":

{ "created": 1717488000, "model": "gemini-3.1-flash-image", "size": "2752x1536", "data": [{ "url": "https://cn.inf.space/api/files/serve/generated-images/....png" }], "usage": { "total_tokens": 1290 } }

Parameters available on this path: model / prompt / size (tier) / aspectRatio / response_format / output_format, plus image / image[] / images for edits. quality, background, and mask have no effect on Gemini. n > 1 is not guaranteed to be fulfilled in one call (see the length of data and the x-image-count-degraded response header for the actual count); for a reliable number of images, make multiple concurrent calls. For descriptions of the common parameters, see Image generation and editing · Common parameters.

Differences from gpt-image-2 / selection guidance

Selection guidance:

• For speed and batch drafts → Small Banana gemini-3.1-flash-image.
• For quality, complex compositions, and commercial posters → Big Banana gemini-3-pro-image.
• For high-concurrency real-time interaction with the lowest latency and unit price, and lightweight edits (stickers / recoloring / background changes) → Small Banana Lite gemini-3.1-flash-lite-image (1K only).
• For exact pixel sizes or mask-based inpainting → prefer gpt-image-2.
• For ultra-wide / ultra-tall banners (1:4, 4:1, 1:8, 8:1) → only Small Banana gemini-3.1-flash-image supports them.
• For long-running jobs (4K / multi-image fusion) → use async jobs.

Pricing

All three models are billed per image at the same price for each size (1K / 2K / 4K); Small Banana Lite outputs 1K only and has a lower unit price. For exact prices, billing rules, and organization-specific pricing, see the Pricing and billing page and the console.