Nano Banana 2 Lite: 4 Seconds, $0.034 an Image

What Nano Banana 2 Lite is and where it fits in the Nano Banana family

Nano Banana 2 Lite is Google's image model released on June 30, 2026, and Google calls it the fastest and most cost-efficient image model in the Nano Banana family so far. Its model ID is gemini-3.1-flash-lite-image, and Google built it for high-throughput work at scale. You may also see it written as Nano Banana Lite.

Google released it on the same day as Gemini Omni Flash, a video model. Google's announcement shows how to chain the two.

Nano Banana 2 Lite pricing: $0.034 per 1K image

Google lists Nano Banana 2 Lite at $0.034 per 1K image. Read "per 1K image" as the price of one image at the 1K size, not a bulk rate. Google pitches it at developers who draft and iterate a lot, run on a tight budget, or work over low bandwidth.

The announcement gives one other price: Gemini Omni Flash video costs $0.10 per second of output. The announcement does not state a price for Nano Banana 2 or Pro.

Speed: about 4 seconds from text to image

Google says Nano Banana 2 Lite delivers text-to-image output in 4 seconds. That figure comes from Google, not from an independent test, so treat it as the vendor's number. Google ties it to interactive prototyping and rapid visual drafting.

The blog gives no time for other jobs, so 4 seconds applies to text-to-image only. Google positions Lite as the family's speed option, built for near-real-time, high-volume work where ultra-low latency is critical.

What it handles well: prompts, consistent characters, readable text

Google names three qualities that Nano Banana 2 Lite keeps despite its focus on speed. All three are Google's claims, not results from an independent test.

• Prompt adherence: the image follows what the prompt asks for.
• Character consistency: a character can stay recognizable from one image to the next.
• Text rendering: words placed inside the image stay legible.

Limits to check before you commit

Google's announcement lists no limitations for Nano Banana 2 Lite, so the limits to plan around come from how Google positions it. Lite is the speed-first model, and Google points to Nano Banana Pro for tasks where accuracy matters more than speed.

Google has published a price for the 1K size only, with no price for any other size. Before an image ships, check three things in the output:

• Any text inside the image, letter by letter.
• Faces and recurring characters, against the earlier images in the set.
• Facts shown in the picture, such as logos, numbers, and product details.

How Lite compares with Nano Banana 2 and Pro: when to step up

Use Lite for drafts and high-volume work, Nano Banana 2 for general use, and Nano Banana Pro when accuracy matters more than speed. Google describes the three models by their purpose, and "How Google describes the three Nano Banana models" follows that wording.

How Google describes the three Nano Banana models

Step up from Lite when the picture is a final asset that needs precise control, or when a mistake in the output costs more than the speed Lite buys. For the Gemini text models, see "Gemini 3.5 Pro is still not in the Gemini API".

Where to use it: Google AI Studio and the Gemini API

Nano Banana 2 Lite is available in Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform, and the model ID to call is gemini-3.1-flash-lite-image. Google's announcement lists all three as available on June 30, 2026. It adds that the model was also rolling out that day in its consumer products, including AI Mode in Search and the Gemini app.

Pairing the image model with Gemini Omni Flash for video

To make a video from a Lite image, generate the still with Nano Banana 2 Lite, pass it to Gemini Omni Flash as a reference, and generate the clip. Gemini Omni Flash is a Gemini model for video generation and conversational editing that entered public preview in Google AI Studio and the Gemini API at launch. Google's announcement lists 10-second videos at $0.10 per second of output, the same rate as Veo 3.1 Fast, so a full clip comes to $1.