Flux.2 [Klein] 9B Distilled: Image Edit

BFL distilled model. Outstanding quality at sub-second speed. Great for real-time generation while retaining quality. Marketing launch will focus on this model.

Download Workflow Try in Cloud

Input

Image to edit

Edit instruction

Megapixels

Advanced settings

Seed

Steps

Default: 4

Run

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": "Flux2-Klein",
            "images": [
                "75:65",
                0
            ]
        },
        "_meta": {
            "title": "Save Image"
        }
    },
    "76": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_FILENAME"
        },
        "_meta": {
            "title": "Load Image"
        }
    },
    "75:61": {
        "class_type": "KSamplerSelect",
        "inputs": {
            "sampler_name": "euler"
        },
        "_meta": {
            "title": "KSamplerSelect"
        }
    },
    "75:62": {
        "class_type": "Flux2Scheduler",
        "inputs": {
            "height": [
                "75:99",
                1
            ],
            "steps": 4,
            "width": [
                "75:99",
                0
            ]
        },
        "_meta": {
            "title": "Flux2Scheduler"
        }
    },
    "75:63": {
        "class_type": "CFGGuider",
        "inputs": {
            "cfg": 1,
            "model": [
                "75:70",
                0
            ],
            "negative": [
                "75:79:100",
                0
            ],
            "positive": [
                "75:79:77",
                0
            ]
        },
        "_meta": {
            "title": "CFG Guider"
        }
    },
    "75:64": {
        "class_type": "SamplerCustomAdvanced",
        "inputs": {
            "guider": [
                "75:63",
                0
            ],
            "latent_image": [
                "75:66",
                0
            ],
            "noise": [
                "75:73",
                0
            ],
            "sampler": [
                "75:61",
                0
            ],
            "sigmas": [
                "75:62",
                0
            ]
        },
        "_meta": {
            "title": "SamplerCustomAdvanced"
        }
    },
    "75:65": {
        "class_type": "VAEDecode",
        "inputs": {
            "samples": [
                "75:64",
                0
            ],
            "vae": [
                "75:72",
                0
            ]
        },
        "_meta": {
            "title": "VAE Decode"
        }
    },
    "75:66": {
        "class_type": "EmptyFlux2LatentImage",
        "inputs": {
            "batch_size": 1,
            "height": [
                "75:99",
                1
            ],
            "width": [
                "75:99",
                0
            ]
        },
        "_meta": {
            "title": "Empty Flux 2 Latent"
        }
    },
    "75:70": {
        "class_type": "UNETLoader",
        "inputs": {
            "unet_name": "flux-2-klein-9b.safetensors",
            "weight_dtype": "default"
        },
        "_meta": {
            "title": "Load Diffusion Model"
        }
    },
    "75:71": {
        "class_type": "CLIPLoader",
        "inputs": {
            "clip_name": "qwen_3_8b_fp8mixed.safetensors",
            "device": "default",
            "type": "flux2"
        },
        "_meta": {
            "title": "Load CLIP"
        }
    },
    "75:72": {
        "class_type": "VAELoader",
        "inputs": {
            "vae_name": "flux2-vae.safetensors"
        },
        "_meta": {
            "title": "Load VAE"
        }
    },
    "75:73": {
        "class_type": "RandomNoise",
        "inputs": {
            "noise_seed": 137120342660015
        },
        "_meta": {
            "title": "RandomNoise"
        }
    },
    "75:74": {
        "class_type": "CLIPTextEncode",
        "inputs": {
            "clip": [
                "75:71",
                0
            ],
            "text": "Turn the amber bottle into a bottle carved from clear blue ice. Keep both hands, their position and grip, the travertine table, the linen runner and the warm plaster wall exactly as they are, with the same camera angle and framing. The ice bottle keeps the same shape, the vertical fluting and the faceted stopper, but it is cold and translucent, with tiny trapped air bubbles, frosted edges and a thin film of meltwater beading on its surface. A few drops have fallen onto the stone and catch the low sunlight. Inside the bottle, the honey-coloured oil now glows a pale mint green. Keep the late-afternoon light from the left and the long soft shadows."
        },
        "_meta": {
            "title": "CLIP Text Encode (Positive Prompt)"
        }
    },
    "75:79:100": {
        "class_type": "ReferenceLatent",
        "inputs": {
            "conditioning": [
                "75:82",
                0
            ],
            "latent": [
                "75:79:78",
                0
            ]
        },
        "_meta": {
            "title": "Set Reference Latent"
        }
    },
    "75:79:77": {
        "class_type": "ReferenceLatent",
        "inputs": {
            "conditioning": [
                "75:74",
                0
            ],
            "latent": [
                "75:79:78",
                0
            ]
        },
        "_meta": {
            "title": "Set Reference Latent"
        }
    },
    "75:79:78": {
        "class_type": "VAEEncode",
        "inputs": {
            "pixels": [
                "75:80",
                0
            ],
            "vae": [
                "75:72",
                0
            ]
        },
        "_meta": {
            "title": "VAE Encode"
        }
    },
    "75:80": {
        "class_type": "ImageScaleToTotalPixels",
        "inputs": {
            "image": [
                "76",
                0
            ],
            "megapixels": 1,
            "resolution_steps": 1,
            "upscale_method": "nearest-exact"
        },
        "_meta": {
            "title": "Scale Image to Total Pixels"
        }
    },
    "75:82": {
        "class_type": "ConditioningZeroOut",
        "inputs": {
            "conditioning": [
                "75:74",
                0
            ]
        },
        "_meta": {
            "title": "Conditioning Zero Out"
        }
    },
    "75:99": {
        "class_type": "GetImageSize",
        "inputs": {
            "image": [
                "75:80",
                0
            ]
        },
        "_meta": {
            "title": "Get Image Size"
        }
    }
})
workflow.set_input("76", "image", client.assets.from_url("https://comfy.org/catalog-media/bccdb702-e6ca-432c-9c0c-1396b7cd4bb9/11f8ca2e-81c9-486d-9dcc-751a717d5a5a"))

# 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 Flux.2 [Klein] 9B Distilled: Image Edit workflow leverages the power of the Flux.2 Klein model to provide high-quality image editing capabilities at impressive speeds. This workflow is designed to enable real-time image generation and editing, making it ideal for applications that require quick turnaround without compromising on quality. The core of this workflow is the Flux.2 Klein model, a distilled version of the original Flux model, which offers outstanding performance by reducing computational load while maintaining the model's efficacy.

Key nodes in this workflow include the SaveImage and LoadImage nodes, which handle the input and output of images, and the MarkdownNote node, which provides documentation and guidance within the workflow itself.

Technically, the workflow operates by taking an input image, processing it through the Flux.2 Klein model using a series of nodes that manage data flow and model inference, and finally outputting an edited image. The 7b34ab90-36f9-45ba-a665-71d418f0df18 node is crucial for interfacing with the model, ensuring that the image data is correctly formatted and processed. This setup makes the workflow highly efficient, allowing users to achieve sub-second image editing, which is particularly beneficial for real-time applications such as live video feeds or interactive design tools.

Flux.2 [Klein] 9B Distilled: Image Edit

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 makes the Flux.2 Klein model suitable for real-time applications?

How do I ensure the best quality output from this workflow?

Can this workflow handle batch processing of images?

Is there any documentation available for understanding subgraphs used in this workflow?