zust-ai/zust-diffusion

auto1111_ds8

Public
61.8K runs

Input

boolean

Check if Server Alive

Default: false

boolean

Detailed Output

Default: false

string
Shift + Return to add a new line

SAM, photoshift, zust_fashion, zust_xl, upscale, enhance, cleanup

Default: "zust_xl"

boolean

PhotoShift (Clean source image subject)

Default: true

string
Shift + Return to add a new line

Photoshift scene image

Default: "{\"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/ps/perfume_scene2.png\"}"

string
Shift + Return to add a new line

Photoshift scene subject(s) image

Default: ""

string
Shift + Return to add a new line

Photoshift subject(s) image

Default: "[{\"left\": 0, \"top\": 0, \"scale\": 0.0009765625, \"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/ps/perfume_raw2.png\"}]"

string
Shift + Return to add a new line

Photoshift prompt

Default: ""

number

Add noise to masked area

Default: 0.65

number

IP-Adapter reference scale

Default: 0.35

number

BLIP Caption strength

Default: 5

boolean

PhotoShift (If resize the scene image)

Default: true

string
Shift + Return to add a new line

Product Tags (jewellery, topdown)

Default: ""

integer

Width

Default: 1024

integer

height

Default: 1024

string
Shift + Return to add a new line

Prompt

Default: "a handbag on a sofa in a sunlit living room with a bookshelf, plush rug, and a framed painting on the wall"

string
Shift + Return to add a new line

Negative Prompt

Default: ""

string
Shift + Return to add a new line

Subject positions & URLs

Default: "[{\"left\": 0, \"top\": 0, \"scale\": 0.0009765625, \"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/zust_product/handbag_square.png\"}]"

boolean

Is this a pro user?

Default: false

number

Controlnet weight for the firstpass

Default: 0.8

number

Noise strength for hires.fix

Default: 0.55

number

Prompt strength for inpaint.fix

Default: 5

string
Shift + Return to add a new line

Subject positions & URLs

Default: "{\"left\": 0, \"top\": 0, \"scale\": 0.0009765625, \"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/zust_fashion/fashion_sq.png\"}"

string
Shift + Return to add a new line

Fashihon pose

Default: "{\"left\": 0, \"top\": 0, \"scale\": 0.0009765625, \"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/zust_fashion/fashion_sq_mask.png\"}"

string
Shift + Return to add a new line

Subject data (this shouldn't change)

Default: "[{\"left\": 0, \"top\": 0, \"scale\": 0.0009765625, \"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/enhance/enhance_sub.png\"}]"

string
Shift + Return to add a new line

Scene to be enhanced

Default: "{\"left\": 0, \"top\": 0, \"scale\": 0.0009765625, \"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/enhance/scene_to_be_enhanced.png\"}"

number

IP-Adapter strength of scene. If 0 no Reference

Default: 0.25

number

Noise to be added in the scene

Default: 0.5

string
Shift + Return to add a new line

Enhanced Prompt

Default: ""

string
Shift + Return to add a new line

Negative Prompt

Default: ""

string
Shift + Return to add a new line

Upscale the Image

Default: "{\"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/ps/perfume_scene2.png\"}"

integer

Upscaling factor

Default: 2

string
Shift + Return to add a new line

Get SAM Embeddings

Default: "{\"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/ps/perfume_scene2.png\"}"

string
Shift + Return to add a new line

Request Type

Default: ""

string
Shift + Return to add a new line

Image URL to be cleaned

Default: "{\"image_url\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/ps/perfume_scene2.png\", \"mask\": \"https://raw.githubusercontent.com/zust-ai/product-ai-training/main/test/ps/perfume_scene_subject2.png\"}"

Output

No output.
Generated in

This output was created using a different version of the model, zust-ai/zust-diffusion:074800d6.

Run time and cost

This model costs approximately $0.055 to run on Replicate, or 18 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia A100 (80GB) GPU hardware. Predictions typically complete within 40 seconds.

Readme

Model description

Intended use

Ethical considerations

Caveats and recommendations