Plain-English Summary
A production image model for generating and editing images while following detailed instructions.
Nano Banana 2
A production image model for generating and editing images while following detailed instructions.
Nano Banana
Image
Stable
Open weights: No
API: Yes
Technical Notes
Gemini image model accepting text, images, and PDFs and producing text plus images up to 4K with a 131K input context.
Best For
Marketing images, product visuals, editing, multi-turn creative workflows
Watch Out For
Image generation can be nondeterministic; visual text and precise layout still need review.
Model Specs
- Maker
- Country
- Not listed
- Family
- Nano Banana
- Type
- Image
- Release date
- Not listed
- Date confidence
- Not publicly stated
- Status
- Stable
- Context window
- 131,072 tokens
- Max output
- 32,768
- Parameters
- Not disclosed
- Active parameters
- Not disclosed
- Architecture
- Proprietary multimodal image model
- Reasoning
- Yes — internal planning
- Tool calling
- No / endpoint-specific
- Structured output
- No
- API available
- Yes
- Open weights
- No
- Self-hostable
- No
- License
- Proprietary
- Inputs
- Text, image, PDF
- Outputs
- Image, text
- Approx. price
- See provider pricing
- How to access
- Provider website or API
- Model / API ID
- gemini-3.1-flash-image
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Prompting / RAG
- Completeness
- 86 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source