Documentation Index
Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
Use this file to discover all available pages before exploring further.
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
Overview
The Krea 2 Image node generates images using the Krea 2 AI model. It supports two model variants: Medium for expressive illustrations and Large for expressive photorealism. You can optionally include a moodboard and up to 10 image style references to influence the generated image.Inputs
| Parameter | Data Type | Required | Range | Description |
|---|---|---|---|---|
prompt | STRING | Yes | N/A | Text prompt for the image. |
model | DICT | Yes | See below | Krea 2 Medium is best for expressive illustrations; Krea 2 Large is best for expressive photorealism. |
seed | INT | Yes | 0 to 2147483647 | Random seed for reproducibility (default: 0). |
model parameter is a dictionary with the following sub-parameters:
| Sub-Parameter | Data Type | Required | Range | Description |
|---|---|---|---|---|
model | STRING | Yes | "krea 2 medium""krea 2 large" | Selects the Krea 2 model variant. |
aspect_ratio | STRING | Yes | N/A | The aspect ratio for the generated image. |
resolution | STRING | Yes | N/A | The resolution for the generated image. |
creativity | FLOAT | Yes | N/A | Controls the creativity level of the generation. |
moodboard_id | STRING | No | N/A | The UUID of a Krea moodboard to influence the image. Must be a valid UUID. |
moodboard_strength | FLOAT | No | N/A | The strength of the moodboard influence (default: 0.35). |
style_reference | LIST | No | 0 to 10 items | A list of image style references. Each reference must have a url (STRING) and strength (FLOAT). |
moodboard_idmust be a valid UUID (e.g.,"123e4567-e89b-12d3-a456-426614174000"). Copy it from the Krea website.style_referenceaccepts a maximum of 10 image style references.- The
promptmust be at least 1 character long.
Outputs
| Output Name | Data Type | Description |
|---|---|---|
image | IMAGE | The generated image as a tensor. |
Source fingerprint (SHA-256):
6aeb2d935ef5df5699a19271c9ceb766892ef4b0e4f67bfa540bf12ffadf362d