Nano Banana 2 Lite

Nano Banana 2 Lite is Google's lightweight image model within the Nano Banana family, carrying the API name gemini-3.1-flash-lite-image. Google released it on June 30, 2026, and explicitly framed it as the fastest and most cost-efficient member of its Gemini image lineup, built for speed and scale rather than maximum creative control. Its position in the family signals that it inherits the multimodal foundations of Google's Gemini image work while being optimized for rapid text-to-image generation and editing, making it the lighter counterpart to Nano Banana 2 standard and Pro for tasks that do not require the most complex image work, multi-reference control, or long multi-turn editing sessions.

Google's stated targets for the model are a roughly 4-second generation time per 1K-resolution image and about $0.034 per 1K image, designed for high-throughput use cases such as ad variants, product mockups, marketplace imagery, thumbnails, UI concepts, and batch creative pipelines. Coursiv describes it as "the model to test when you need lots of good-enough images quickly," and notes availability through Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform, alongside consumer surfaces including AI Mode in Search, the Gemini app, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads. Practical fit centers on prototyping and bulk generation, while users needing deeper reasoning or richer compositional control are pointed toward Nano Banana 2 or Nano Banana Pro.

Kilo Gatewaygoogle/gemini-3.1-flash-lite-imagegemini-flash-lite

Open provider page Provider docs

Quick Info

Provider: Kilo Gateway
Model key: google/gemini-3.1-flash-lite-image
Release date: Jun 30, 2026
Last updated: Jun 30, 2026
Knowledge cutoff: 2025-01
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $0.25
Output token cost: $1.50

Limits

Output tokens: 58,982 tokens
Context window: 65,536 tokens

Transparent token rates

Compare Nano Banana 2 Lite pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Nano Banana 2 Lite

Kilo Gateway

Coverage

The Qatar News Agency wire, dated July 1, 2026, paraphrases Google's announcement and identifies the model both as Nano Banana 2 Lite and by its alternate name Gemini 3.1 Flash-Lite Image, confirming it is Google's fastest and most cost-efficient image generation model. It states the model generates a 1K-resolution ima The piece specifies that Nano Banana 2 Lite is built on the Gemini 3.1 Flash-Lite architecture and is capped at 1K resolution, unlike Nano Banana 2 and Nano Banana Pro which support up to 4K — a deliberate tradeoff that contributes to its lower latency and cost. It also lists enhancements over its predecessor Nano Bana

Kilo Gateway

Coverage

The 36Kr piece explicitly identifies the model by its official API name gemini-3.1-flash-lite-image, matching the subject modelKey, and confirms the core spec sheet: 1K images generated in approximately 4 seconds at $0.034 per 1K-resolution image. It also lists the three availability surfaces — Google AI Studio, the Ge The article frames Nano Banana 2 Lite as a direct competitor to ByteDance's Seedream 5.0/5.1 Lite, reporting a Text-to-Image Elo of 1251 for NB2 Lite versus 1132 for Seedream 5.0 Lite, and an end-to-end latency of roughly 4.0 seconds. Pricing is essentially matched at $0.034 vs ~$0.035 per image, so the piece frames NB

Kilo Gateway

Coverage

Google released Nano Banana 2 Lite on June 30, 2026, as the fastest and most affordable member of its Nano Banana image generator family, succeeding the original Nano Banana (powered by Gemini 3.1 Flash) and the February 2026 Nano Banana 2 release. The model is positioned as a "generalist workhorse"-adjacent option opt Beyond raw speed and cost, the launch positions Nano Banana 2 Lite for enterprise and developer pipelines that need throughput over maximum fidelity, contrasting with Nano Banana 2 and Nano Banana Pro, which target higher-quality outputs. The TechCrunch excerpt frames the release within Google's broader competitive pus

Kilo Gateway

CoverageBenchmark

K-Dense ran an independent benchmark of Nano Banana 2 Lite on scientific figure generation, generating 240 figures across four image models, 20 prompts, and three samples per prompt, scored blind against the original prompt on a five-part rubric. The headline result is that Nano Banana 2 Lite returns a 1K scientific fi Under the benchmark's explicit overall score weighting 70% quality and 30% speed, Nano Banana 2 Lite ranks first with an overall score of 76.9, meaning its latency advantage is large enough to shift the product tradeoff when scientific figure generation is a repeated tool call inside a research session. The piece frame