Nano Banana 2 Lite API: what Google shipped and what Sume lists
Nano Banana 2 Lite is Google's gemini-3.1-flash-lite-image model. What Google says it does, and which Nano Banana models Sume's image API lists today.

Nano Banana 2 Lite is Google's gemini-3.1-flash-lite-image model, released on 2026-06-30 in the Gemini API and Google AI Studio. Sume's image catalog does not currently list it: the Nano Banana models Sume lists are google/nano-banana-2 and google/nano-banana-pro, so a request naming Lite will not resolve there.
Google's side comes from its launch post and Gemini API docs; Sume's side from the Image API docs and the image catalog code. All read 2026-09-29.
What is Nano Banana 2 Lite?
Google describes it as a speed- and cost-focused Gemini image model for near-real-time, high-volume work, with prompt adherence, character consistency and legible in-image text. Google's launch post says text-to-image outputs arrive in 4 seconds, and recommends it as the replacement for gemini-2.5-flash-image.
It is one of four Nano Banana models in Google's docs, next to gemini-3.1-flash-image, gemini-3-pro-image and the legacy gemini-2.5-flash-image.
What does Google's documentation say about its limits?
| Item | Google documents |
|---|---|
| Model id | gemini-3.1-flash-lite-image |
| Release | 2026-06-30 |
| Price | $0.034 per 1K image (launch post); $0.0336 standard (pricing page) |
| Resolution | 1K only |
| Reference images | Up to 14 object images; no character-consistency image slots |
| Watermark | All generated images include a SynthID watermark |
Can I call Nano Banana 2 Lite through Sume?
Not as a listed model today. POST /v1/images accepts the ids in GET /v1/images/models, and a model that is not in that catalog is not accepted. The catalog is the source of truth, so check it instead of trusting a blog post, including this one:
curl "https://api.sume.com/v1/images/models" \
-H "Authorization: Bearer $SUME_API_KEY"Which Nano Banana models can I use on Sume now?
google/nano-banana-2 and google/nano-banana-pro both take a resolution tier from 512 to 4K, a native aspect_ratio, and up to 10 input_references. One 1K Nano Banana 2 image is estimated at $0.10 on Sume. Model-by-model detail is in Nano Banana Pro API on Sume.
curl -X POST https://api.sume.com/v1/images \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google/nano-banana-2",
"prompt": "A ceramic mug on a linen cloth, soft window light",
"resolution": "1K",
"aspect_ratio": "1:1"
}'Will Sume add it?
The docs make no promise either way, so this post makes none. When a model reaches the catalog it shows up in GET /v1/images/models with its supported parameters and its pricing line.
Sources
Related posts
More in Models
- Nano Banana Pro vs Nano Banana 2: which id to send on Sume
Nano Banana Pro and Nano Banana 2 share tiers and 10 references on Sume; they differ in extra aspect ratios and in which tiers change the price.
- Open-source video model vs API: which to use for Wan-class video
Weights you run yourself, or a hosted video API? What the choice changes for cost, setup and limits, with Wan 3.0's per-second API price as the worked example.
- Remove shadow from photo with AI: edit it or cut it out
To remove a shadow from a photo with AI, send the photo to an image-edit model, name the one shadow to remove, and list what stays. Prompts, cost, limits.
- Seedance 2.5 aspect ratios: 9:16 vertical, 21:9 and the rest
Seedance 2.5 on Sume accepts 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16. Set aspect_ratio explicitly; 3:2 is rejected. Ratios, errors and the price effect.
Written by Sume