NVIDIA Models
OpenRouter
NVIDIA TTS
Flux.1 POC
Generate an image via the NVIDIA API and download the result.
Prompt
Model
Flux.1 Dev
Flux.1 Kontext Dev
Flux.1 Schnell
Stable Diffusion 3 Medium
Different models support different inputs; the form adapts based on selection.
Mode
base
canny
depth
base is text-to-image; canny/depth use an input image.
Negative Prompt
Shown for models that support it.
Aspect Ratio
Used by some models (e.g. SD3, Flux Kontext).
CFG Scale
Width
Height
Seed
Steps
Preprocess Image
true
false
Used for canny/depth.
Input Image (optional)
None
Upload
example_id
Upload Image
Will be sent as a data URL in JSON.
example_id
Sent as
data:image/png;example_id,<id>
Advanced (JSON overrides)
Extra JSON
This must be a JSON object. It will be merged into the payload sent to the model (prompt from the form always wins).
Generate
Result
Raw JSON