Nano Banana 2 is live on AI Gateway
Google's Nano Banana 2 now live on AI Gateway—faster image generation with web-grounded outputs and configurable reasoning levels.
Why it matters
Vercel's AI Gateway adds Google's latest multimodal model with real-time image search grounding and configurable thinking—expanding the unified API's model roster and enabling developers to balance speed, cost, and reasoning depth for image generation workloads.
The key facts
13 to knowGemini 3.1 Flash Image Preview (Nano Banana 2) released on Vercel AI Gateway
Model integrates Google Image Search for real-world visual grounding
Configurable thinking levels: Minimal and High for reasoning complexity
New resolutions and aspect ratios: 512p with 1:4 and 1:8 ratios
Multimodal—supports both text and image generation in single API call
Available via AI SDK with streamText and generateText methods
AI Gateway provides unified API, usage tracking, cost monitoring, and intelligent provider routing
Gemini 3.1 Flash Image Preview (Nano Banana 2) launched on AI Gateway
Supports new resolutions and aspect ratios: 512p, 1:4, 1:8
Includes Google Image Search grounding for real-world imagery retrieval
Configurable thinking levels: Minimal and High for variable reasoning depth
Available via Vercel AI SDK with streamText/generateText for multimodal generation
AI Gateway provides unified API, cost tracking, observability, and intelligent provider routing
Go to the source
Vercel Blogvercel.com
Publisher excerpt: Gemini 3.1 Flash Image Preview (Nano Banana 2) is now available on AI Gateway. This release improves visual quality while maintaining the generation speed and cost of flash-tier models. Nano Banana 2 can use Google Image Search to ground outputs in real-world imagery. This helps with rendering…