Image Upscale: Z-Image-Turbo 2K

Upload an image to upscale it to 2K resolution using the Z-Image-Turbo model.

Output

Example

Call this model from your code

The request below uses the same prepared graph and settings as the Playground. Each submission can start a paid run; do not automatically retry a submission whose outcome is unknown.

Get API key

What you need

Runs on: Comfy Cloud
Key: COMFY_API_KEY
Your files: Uploaded by the code from their urls

Needs a paid Cloud plan and credits.

Download API JSON

# Python 3.10+: pip install comfy-sdk==0.4.0
import os

from comfy_sdk import Comfy

api_key = os.environ.get("COMFY_API_KEY", "").strip()
if not api_key: raise RuntimeError("Set COMFY_API_KEY")
os.environ.setdefault("COMFY_BASE_URL", "https://cloud.comfy.org")
client = Comfy(api_key=api_key)
workflow = client.workflows.from_json({
    "9": {
        "class_type": "SaveImage",
        "inputs": {
            "filename_prefix": "z-image-upscaled",
            "images": [
                "87:65",
                0
            ]
        },
        "_meta": {
            "title": "Save Image"
        }
    },
    "77": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_FILENAME"
        },
        "_meta": {
            "title": "Load Image"
        }
    },
    "86": {
        "class_type": "ImageCompare",
        "inputs": {
            "compare_view": {
                "__value__": [
                    "",
                    ""
                ]
            },
            "image_a": [
                "77",
                0
            ],
            "image_b": [
                "87:65",
                0
            ]
        },
        "_meta": {
            "title": "Compare Images"
        }
    },
    "87:62": {
        "class_type": "CLIPLoader",
        "inputs": {
            "clip_name": "qwen_3_4b.safetensors",
            "device": "default",
            "type": "lumina2"
        },
        "_meta": {
            "title": "Load CLIP"
        }
    },
    "87:63": {
        "class_type": "VAELoader",
        "inputs": {
            "vae_name": "ae.safetensors"
        },
        "_meta": {
            "title": "Load VAE"
        }
    },
    "87:65": {
        "class_type": "VAEDecode",
        "inputs": {
            "samples": [
                "87:69",
                0
            ],
            "vae": [
                "87:63",
                0
            ]
        },
        "_meta": {
            "title": "VAE Decode"
        }
    },
    "87:66": {
        "class_type": "UNETLoader",
        "inputs": {
            "unet_name": "z_image_turbo_bf16.safetensors",
            "weight_dtype": "default"
        },
        "_meta": {
            "title": "Load Diffusion Model"
        }
    },
    "87:67": {
        "class_type": "CLIPTextEncode",
        "inputs": {
            "clip": [
                "87:62",
                0
            ],
            "text": "masterpiece, 8k"
        },
        "_meta": {
            "title": "CLIP Text Encode (Positive Prompt)"
        }
    },
    "87:69": {
        "class_type": "KSampler",
        "inputs": {
            "cfg": 1,
            "denoise": 0.33,
            "latent_image": [
                "87:80",
                0
            ],
            "model": [
                "87:70",
                0
            ],
            "negative": [
                "87:71",
                0
            ],
            "positive": [
                "87:67",
                0
            ],
            "sampler_name": "dpmpp_2m_sde",
            "scheduler": "beta",
            "seed": 824287194145573,
            "steps": 5
        },
        "_meta": {
            "title": "KSampler"
        }
    },
    "87:70": {
        "class_type": "ModelSamplingAuraFlow",
        "inputs": {
            "model": [
                "87:66",
                0
            ],
            "sampling": "flow",
            "shift": 3
        },
        "_meta": {
            "title": "ModelSamplingAuraFlow"
        }
    },
    "87:71": {
        "class_type": "CLIPTextEncode",
        "inputs": {
            "clip": [
                "87:62",
                0
            ],
            "text": ""
        },
        "_meta": {
            "title": "CLIP Text Encode (Negative Prompt)"
        }
    },
    "87:76": {
        "class_type": "UpscaleModelLoader",
        "inputs": {
            "model_name": "RealESRGAN_x4plus.safetensors"
        },
        "_meta": {
            "title": "Load Upscale Model"
        }
    },
    "87:78": {
        "class_type": "ImageScaleToTotalPixels",
        "inputs": {
            "image": [
                "77",
                0
            ],
            "megapixels": 1,
            "resolution_steps": 1,
            "upscale_method": "lanczos"
        },
        "_meta": {
            "title": "Scale Image to Total Pixels"
        }
    },
    "87:79": {
        "class_type": "ImageUpscaleWithModel",
        "inputs": {
            "image": [
                "87:78",
                0
            ],
            "upscale_model": [
                "87:76",
                0
            ]
        },
        "_meta": {
            "title": "Upscale Image (using Model)"
        }
    },
    "87:80": {
        "class_type": "VAEEncode",
        "inputs": {
            "pixels": [
                "87:81",
                0
            ],
            "vae": [
                "87:63",
                0
            ]
        },
        "_meta": {
            "title": "VAE Encode"
        }
    },
    "87:81": {
        "class_type": "ImageScaleBy",
        "inputs": {
            "image": [
                "87:79",
                0
            ],
            "scale_by": 0.5,
            "upscale_method": "lanczos"
        },
        "_meta": {
            "title": "Upscale Image By"
        }
    }
})
workflow.set_input("77", "image", client.assets.from_url("https://comfy.org/catalog-media/b6246849-5e0a-4dd7-bd25-e2c86e5f8a06/b21718bf-6c2f-48b3-95bb-ac9dd2444bc2"))

# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)
for output in job.get_outputs("9"):
    output.to_file(output.name)

API documentation

Try an example

About this workflow

The 'Image Upscale: Z-Image-Turbo 2K' workflow is designed to upscale images to a 2K resolution using the advanced Z-Image-Turbo model. This workflow is particularly useful for enhancing the quality of lower-resolution images, making them suitable for high-quality displays or print. The process begins with the 'LoadImage' node, which allows users to upload their initial image. The core of the workflow is the 'dd15cfd3-cd53-428c-b3e2-33ed4ff8fa78' node, which utilizes the Z-Image-Turbo model to upscale the image, applying sophisticated algorithms to maintain and enhance detail.

The workflow also includes a 'SaveImage' node to store the upscaled image and an 'ImageCompare' node to visually compare the original and upscaled images, ensuring quality control. A 'MarkdownNote' node provides users with insights into the denoise parameter, which controls the model's creative interpretation during upscaling. Lower denoise values keep the output closer to the original, while higher values allow for more creative reinterpretation, adding details and potentially altering the image's appearance.

This flexibility makes the workflow adaptable to various needs, from subtle enhancements to more artistic transformations.

Image Upscale: Z-Image-Turbo 2K

Input

Inside the workflow

Call this model from your code

What you need

Try an example

About this workflow

Questions & answers

Input

Inside the workflow

Call this model from your code

What you need

Try an example

About this workflow

Questions & answers

What is the purpose of the denoise parameter?

Can this workflow handle batch processing of images?

Is the upscaled image quality suitable for professional printing?

How does the ImageCompare node enhance the workflow?