ComfyUI-AtlasCloud

ComfyUI AtlasCloud

Image generation nodes for atlascloud.ai, powered by OpenAI's GPT Image 2 models.

Features

• Text to Image — generate images from text descriptions
• Image to Image — edit/transform images using up to 3 reference images
• Full support for GPT Image 2 parameters: quality, size, output format, seed
• Configurable base URL for self-hosted or alternative API endpoints
• Extra JSON params for advanced/custom API features

AtlasCloud Nodes

Requirements

• ComfyUI
• Python packages: torch, pillow, numpy, requests

Installation

cd ComfyUI/custom_nodes/
git clone https://github.com/clownvary/ComfyUI-AtlasCloud.git

Restart ComfyUI (or reload custom nodes). The nodes will appear under the AtlasCloud category.

Nodes

AtlasCloud Text to Image

Text-to-image generation using GPT Image 2.

Parameter	Type	Default	Description
base_url	STRING	https://api.atlascloud.ai	API base URL
api_key	STRING	—	AtlasCloud API key (required)
prompt	STRING	—	Text description of the image to generate (required)
model	STRING	openai/gpt-image-2/text-to-image	AI model to use
size	ENUM	1024x1024	Output resolution (width x height)
quality	ENUM	medium	Generation quality: low, medium, high
output_format	ENUM	jpeg	Output image format: jpeg, png
seed	INT	-1	Random seed for reproducibility (-1 = random)
extra_params	STRING	—	Additional JSON parameters merged into the API request body

Outputs: IMAGE (tensor), STRING (info JSON)

---

AtlasCloud Image to Image

Image editing/transformation using up to 3 reference images.

Parameter	Type	Default	Description
base_url	STRING	https://api.atlascloud.ai	API base URL
api_key	STRING	—	AtlasCloud API key (required)
image_1	IMAGE	—	Primary reference image (required)
prompt	STRING	—	Description of how to edit the image(s) (required)
image_2	IMAGE	—	Secondary reference image (optional)
image_3	IMAGE	—	Tertiary reference image (optional)
model	STRING	openai/gpt-image-2/edit	AI model to use
size	ENUM	1024x1024	Output resolution (width x height)
quality	ENUM	medium	Generation quality: low, medium, high
output_format	ENUM	jpeg	Output image format: jpeg, png
input_fidelity	ENUM	high	Detail preservation: high (keep faces/logos), low (more creative freedom)
seed	INT	-1	Random seed for reproducibility (-1 = random)
extra_params	STRING	—	Additional JSON parameters merged into the API request body

Outputs: IMAGE (tensor), STRING (info JSON)

Supported Resolutions

Tier	Resolutions
1K	1024x768, 768x1024, 1024x1024, 1024x1536, 1536x1024
2K	1920x1080, 1080x1920, 2560x1440, 1440x2560
3K	3840x2160, 2160x3840

License

MIT