HedraImage to imageComing soon

nano-banana-2

Name: nano-banana-2
Brand: Hedra

nano-banana-2 is an image to image tool from Hedra that runs in the background and returns images. Coming soon: listed with its price, not yet open to runs.

This tool is listed but not yet available to run.

Price pending

Pricing

Price pending.

Limits

Runs: In the background
Returns: image
Longest queue wait: 3600s
Run deadline: 1800s
Largest inline input: 1024 KB
Largest input file: 500 MB
Largest output: 4768 MB
Outputs per run: 16

Input

Field	Kind	Default	Description
`prompt` required	text	—	Generation prompt.
`resolution` required	one of 512px, 1K, 2K, 4K	—	Output resolution.
`aspect_ratio` required	one of adaptive, 16:9, 9:16, 1:1, 21:9, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 1:4, 4:1, 1:8, 8:1	—	Output aspect ratio. 'adaptive' lets the model size the output itself — matching the source image when you pass one.
`seed`	number (whole)	—	Seed for reproducible output; omit for a random seed.
`images`	JSON	—	Images to edit or blend, at most 10 high-fidelity objects and 4 characters. 1 to 14 images, each at most 30 MB.
`num_outputs`	one of 1	1	Number of outputs generated per job. Only 1 is supported.
`image_search`	true or false	false	Let the grounding search return images as well as text, so the model sees what it found rather than only reading about it. Turning this on grounds the generation whether or not google_search is also set.
`google_search`	true or false	false	Ground the generation in live Google Search results, so a prompt about current events or real-world specifics draws on what the web says now. Grounded generations cost more than ungrounded ones.
`enhance_prompt`	true or false	false	Rewrite the prompt before generation. An LLM expands it into a fuller description and the model receives that text instead of the submitted one; the result's `prompt` reports what ran.
`thinking_level`	one of minimal, high	—	How much the model plans before it draws. Omit for the model's own default ('minimal'); 'high' reasons further at the cost of latency.

Same request in code

curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "hedra",
    "model": "nano-banana-2",
    "input": {
      "prompt": "<prompt>",
      "resolution": "512px",
      "aspect_ratio": "adaptive"
    }
  }'