Nano Banana 2
#18 in Text-to-ImageUnknown · v2 · nano · since 26. Februar 2026 · 2× · last seen Aug 10, 2026
21
Momentum
Nano Banana 2 (official technical name: Gemini 3.1 Flash Image) is Google's image generation and editing model that combines Flash-tier speed with Nano Banana Pro-level image quality. It supports text-to-image generation and conversational image editing with resolutions from 0.5K to 4K, improved text rendering, and real-world knowledge via Google Search grounding. The model was announced on February 26, 2026, and is available via Google AI Studio, the Gemini API, and the Gemini app. Billing is handled through Google Cloud/Gemini API based on output image resolution (token-based).
Momentum trend
12.05.10.08.
Features
| Fine-Tuning | No fine-tuning; instead prompt-based control, reference images (up to 14), and subject/character consistency for up to 5 people |
| Generation Time | Approx. 4–15 seconds at 1K, approx. 10–56 seconds at 4K (real-world testing) |
| License | Proprietary Google model; documentation content under Creative Commons Attribution 4.0, code samples under Apache 2.0 |
| Max Resolution | 4K (up to approx. 16 megapixels); also supports 0.5K/1K/2K |
| Platform | Google AI Studio, Gemini API, Gemini app, Google Search, Google Ads, Vertex AI/Gemini Enterprise Agent Platform |
| Price | API: $0.045 (0.5K), $0.067 (1K), $0.101 (2K), $0.151 (4K) per image; Batch mode 50% cheaper |
| Release Date | February 26, 2026 (announcement and rollout in the Gemini app) |