Nano Banana 2 Lite
The fastest, most cost-efficient image generation model in the Nano Banana family
Context window
1M
Input / 1M tokens
0.25
Output / 1M tokens
1.50
Provider
Data verified 2026-07-15
Nano Banana 2 Lite is Google's entry-level image generation model optimized for speed and cost efficiency. It generates text-to-image outputs in approximately 4 seconds at $0.034 per 1,000 images, positioning it as the lowest-cost option in Google's image model family. Designed for high-throughput, rapid iteration workflows where speed and cost matter more than advanced reasoning.
Capability index
Relative estimates (0-100) to place this model against its peers, grounded in published benchmarks.
How to access it
Available through Google AI Studio, Gemini API, and Gemini Enterprise Agent Platform. Rolling out to consumer surfaces including AI Mode in Search, Gemini app, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads.
Strengths
- ✓Ultra-fast 4-second generation time
- ✓Lowest cost in Google's image lineup
- ✓High throughput for batch generation
- ✓Strong character consistency
- ✓Legible in-image text rendering
Best for developers who...
When to choose it (and when not to)
Reach for Nano Banana 2 Lite when...
- →When cost-per-sample is the primary constraint
- →For workflows requiring 100+ images per session
- →When generation latency must be under 5 seconds
- →For rapid prototyping and exploration
Look elsewhere if...
- ✕Use cases requiring advanced reasoning or complex scene composition
- ✕When legible signage, infographic data, or narrative continuity is critical
- ✕Projects where maximum fidelity and detail are essential
- ✕Applications needing search grounding for real-world context
How to use it
- ›Use concise, direct prompts for optimal results
- ›Include style references in descriptions
- ›Specify desired aspect ratios explicitly
- ›Chain with Gemini Omni Flash for image-to-video workflows
Quickstart
Pythonimport google.generativeai as genai
genai.configure(api_key='YOUR_API_KEY')
model = genai.ImageGenerationModel('gemini-3.1-flash-lite-image')
response = model.generate_images(prompt='A sunset over mountains')Visit Google AI Studio for playground access without API key
API model id: gemini-3.1-flash-lite-image
Benchmarks
| Benchmark | Score | Notes |
|---|---|---|
| Arena.ai Text-to-Image Leaderboard | Ranked #5 | Positioned almost as highly as non-Lite Nano Banana 2 despite lower latency |
| Text-to-Image Elo Score | 1251 | Exceeds Nano Banana Pro (1245) |
| Single-Image Editing Elo Score | 1308 | |
| Multiple-Image Editing Elo Score | 1294 | |
| Text-to-Image Arena Elo | 1251 | Beats Nano Banana Pro (1245) and original Nano Banana (1151) |
| Generation Speed | ~4 seconds | Text-to-image generation in ~4 seconds |
| Arena.ai Elo (Image Generation) | 1251 | Text-to-image quality ranking; comparable to Nano Banana 2 |
| Text-to-image latency | ~4 seconds | Standard mode (low-thinking default) |
| Arena.ai Elo (Text-to-Image) | 1251 | Higher than Nano Banana Pro (1245) |
| Generation Latency | ~4 seconds | Text-to-image generation time at 1K resolution |
Source: Arena.ai
Compare Nano Banana 2 Lite
Compare Nano Banana 2 Lite with any other model
Build a comparison →All model comparisons →