model release

Google launches Gemini 3.1 Flash Lite Image with 4-second generation time, $0.25 per 1M input tokens

TL;DR

Google has released Gemini 3.1 Flash Lite Image, a text-to-image model that generates 1K resolution images in approximately 4 seconds — 2.7× faster than Gemini 3.1 Flash Image. The model is priced at $0.25 per 1M input tokens and $1.50 per 1M output tokens, with a 66K context window and knowledge cutoff of January 2025.

2 min read
0

Google launches Gemini 3.1 Flash Lite Image with 4-second generation time

Google has released Gemini 3.1 Flash Lite Image, marketed internally as "Nano Banana 2 Lite," a text-to-image model designed for high-velocity developer pipelines. According to Google, the model generates images in approximately 4 seconds, making it 2.7× faster than Gemini 3.1 Flash Image.

Pricing and specifications

  • Input cost: $0.25 per 1M tokens
  • Output cost: $1.50 per 1M tokens
  • Context window: 66K tokens
  • Knowledge cutoff: January 2025
  • Release date: June 30, 2026 (according to listing)
  • Output resolution: 1K across 14 aspect ratios

Core capabilities

The model supports three primary functions through a single API endpoint: text-to-image generation, image editing, and multi-image composition. As a multimodal model, it returns both text and image outputs.

Google claims the model maintains "character consistency, precise editing, and real-world knowledge" from the Nano Banana family while optimizing for speed and cost efficiency.

All outputs include an invisible SynthID watermark for AI-generated content identification.

Technical positioning

Google positions this model as the "best balance of quality and speed" in the Nano Banana 2 line. The company states it's designed for prototyping, real-time applications, and visual workflows that require generating thousands of images at scale.

The model is available through OpenRouter's API infrastructure, which routes requests across multiple hosting providers based on price, speed, or specific provider requirements. OpenRouter's data indicates customers achieve 60-80% lower effective costs than list prices when using prompt caching with repeated context.

What this means

The 4-second generation time and sub-$0.30 input pricing puts this firmly in the rapid prototyping and high-volume application space. The 66K context window is unusually large for an image model, suggesting it can handle extensive text instructions or multi-turn editing conversations without losing context. However, the June 2026 release date listed appears inconsistent with current timeline — this may be a data error or placeholder value that requires clarification from Google.

Related Articles

model release

Meta Releases Muse Glimmer 30B, an Open-Weight Agentic Model for Consumer Hardware

Meta Superintelligence Labs has released Muse Glimmer 30B, a dense open-weight model distilled from its larger Muse Spark system and tuned for agentic workflows on consumer hardware. The model supports 131K context, image understanding, and over 100 languages at $0.30/$1.10 per 1M input/output tokens.

model release

Google Launches Gemini 3.8 Flash TTS: Voice Cloning and Text-Described Voices for $9-18 per Million Audio Tokens

Google has released Gemini 3.8 Flash TTS and Flash-Lite TTS, two speech generation models that let users design voices from text descriptions or clone a voice from a 30-second sample. Both support over 100 languages and roll out now through the Gemini API and Google AI Studio.

model release

Google Launches Gemini 3.8 Flash TTS and Flash-Lite TTS with Voice Creation from Text Prompts

Google DeepMind has released Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, text-to-speech models that generate custom voices from natural language prompts and support line-by-line performance direction. The models top Hume AI's Voice Design Benchmark at 71.4 and claim first and second place on its Overall Quality Index.

model release

Google DeepMind's New Chief Prioritizes Fast Gemini 4 Release Over AGI Debate

Google DeepMind's new head Koray Kavukcuoglu says Gemini 4 is in early post-training and could ship well before year-end, following the quiet cancellation of Gemini 3.5 Pro. He downplayed the AGI question that defined predecessor Demis Hassabis's tenure, calling it 'not the right conversation.'

Comments

Loading...