Google's Nano Banana 2, internally labelled as Gemini 3.1 Flash Image Preview, is a multimodal model supporting text and image input with image output capability. This makes it unusual in the field as a model that can generate images rather than just analyse them. However, no benchmark data exists for reasoning, coding, or instruction-following performance, so its general intelligence cannot be assessed.

The 131K context window is smaller than most current competitors, and at $0.50/1M input and $3.00/1M output it is priced above lightweight alternatives without the benchmark evidence to justify the premium. The image generation capability is the primary differentiator, which could be valuable for creative or content production workflows.

Until benchmark data is available, this model is best treated as an experimental tool for image-generation use cases rather than a general-purpose business model. Teams with specific image output requirements may find it worth piloting, but broader adoption should wait for validated performance data.

Assessed June 6, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?