v1/jobs

POST 
/v1/jobs

create diffusion job

Request

application/json

Body

required

requestId stringrequired

ensure request idempotence, should be unique

stages object[]required
stages to be executed
Array [
type stage type (string)required
Possible values: [DEFAULT, INPUT_INITIALIZE, DIFFUSION, IMAGE_TO_UPSCALER, IMAGE_TO_ADETAILER, IMAGE_TO_INPAINT, IMAGE_TO_ANIMATE_DIFF]
Default value: DEFAULT
stage type
inputInitialize object
seed int64
Possible values: <= 4294967295
Default value: 0
Random noise seed (omit this option or use 0 for a random seed).
imageResourceId string
Image used to initialize the diffusion process, in lieu of random noise.
count int32
Possible values: >= 1 and <= 4
Default value: 1
Number of images to generate
diffusion object
width int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
height int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
prompts object[]required
Possible values: <= 150
An array of text prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
negativePrompts object[]required
Possible values: <= 150
An array of text negative prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
sdModel string
The model to use for the diffusion , How to get the model id
sdVae string
The vae to use for the diffusion Support list
sampler string
Which sampler to use for the diffusion process. If this value is omitted we'll automatically select an appropriate sampler for you. Support list
steps int32
Possible values: >= 1 and <= 60
Default value: 0
Number of diffusion steps to run.
cfgScale float
Possible values: <= 30
Default value: 7
How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt)
clipSkip int32
denoisingStrength float
etaNoiseSeedDelta int32
controlnet object
args object[]
Array [
inputImageResourceId string
maskResourceId string
preprocessor stringrequired
The model to use for the controlnet preprocessor Support list
model stringrequired
The model to use for the controlnet Support list
weight float
resizeMode string
Possible values: [DEFAULT, JUST_RESIZE, CROP_AND_RESIZE, RESIZE_AND_FILL]
Default value: DEFAULT
guidance float
guidanceStart float
guidanceEnd float
controlMode string
Possible values: [DEFAULT, BALANCED, MY_PROMPT_IS_MORE_IMPORTANT, CONTROLNET_IS_MORE_IMPORTANT]
Default value: DEFAULT
pixelPerfect boolean
preprocessorParams object
]
lora object
items object[]
Array [
loraModel string
The model to use for the diffusion , How to get the model id
weight float
blockWeight string
lora block weight, value such as <weight>:lbw=<layer weight> example: "1:lbw=1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0"
loraAccessKey string
]
animateDiff object
args object[]
Array [
videoLength int32
fps int32
]
embedding object
items object[]
Array [
model string
The model to use for the negative prompts embedding , How to get the model id
weight float
embeddingAccessKey string
]
v1Clip boolean
modelAccessKey string
scheduleName string
enableElla todo：优化成stages,因为底层还不支持，所以目前优化没用，暂时只在推理部分支持。 (boolean)
enablePix2pix boolean
useHunyuanDit deprecated (boolean)
layerDiffusion object
enable boolean
weight float
clipEncoderName string
imageToUpscaler object
hrUpscaler stringrequired
The model to use for the upscaling Support list
hrResizeX int64
Possible values: >= 128 and <= 5120
hr_scale or hr_resize_x must be specified. If hr_scale is specified, hr_resize_x will be ignored. Height of the image upscaler in pixels. Must be in increments of 64 and pass the following validation:
• 262,144 ≤ hr_resize_x * hr_resize_y ≤ 8,294,400
hrResizeY int64
Possible values: >= 128 and <= 5120
hr_scale or hr_resize_y must be specified. If hr_scale is specified, hr_resize_y will be ignored. Height of the image upscaler in pixels. Must be in increments of 64 and pass the following validation:
• 262,144 ≤ hr_resize_x * hr_resize_y ≤ 8,294,400
hrScale double
The size to use for the upscaling
• 262,144 ≤ hr_resize_x * hr_resize_y ≤ 8,294,400
hrSecondPassSteps int32required
Possible values: <= 60
Number of diffusion steps to run.
denoisingStrength floatrequired
Possible values: <= 1
denoising_strength
diffusion object
if has diffusion stage, this diffusion will be ignored, else need to be specified
width int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
height int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
prompts object[]required
Possible values: <= 150
An array of text prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
negativePrompts object[]required
Possible values: <= 150
An array of text negative prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
sdModel string
The model to use for the diffusion , How to get the model id
sdVae string
The vae to use for the diffusion Support list
sampler string
Which sampler to use for the diffusion process. If this value is omitted we'll automatically select an appropriate sampler for you. Support list
steps int32
Possible values: >= 1 and <= 60
Default value: 0
Number of diffusion steps to run.
cfgScale float
Possible values: <= 30
Default value: 7
How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt)
clipSkip int32
denoisingStrength float
etaNoiseSeedDelta int32
controlnet object
args object[]
Array [
inputImageResourceId string
maskResourceId string
preprocessor stringrequired
The model to use for the controlnet preprocessor Support list
model stringrequired
The model to use for the controlnet Support list
weight float
resizeMode string
Possible values: [DEFAULT, JUST_RESIZE, CROP_AND_RESIZE, RESIZE_AND_FILL]
Default value: DEFAULT
guidance float
guidanceStart float
guidanceEnd float
controlMode string
Possible values: [DEFAULT, BALANCED, MY_PROMPT_IS_MORE_IMPORTANT, CONTROLNET_IS_MORE_IMPORTANT]
Default value: DEFAULT
pixelPerfect boolean
preprocessorParams object
]
lora object
items object[]
Array [
loraModel string
The model to use for the diffusion , How to get the model id
weight float
blockWeight string
lora block weight, value such as <weight>:lbw=<layer weight> example: "1:lbw=1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0"
loraAccessKey string
]
animateDiff object
args object[]
Array [
videoLength int32
fps int32
]
embedding object
items object[]
Array [
model string
The model to use for the negative prompts embedding , How to get the model id
weight float
embeddingAccessKey string
]
v1Clip boolean
modelAccessKey string
scheduleName string
enableElla todo：优化成stages,因为底层还不支持，所以目前优化没用，暂时只在推理部分支持。 (boolean)
enablePix2pix boolean
useHunyuanDit deprecated (boolean)
layerDiffusion object
enable boolean
weight float
clipEncoderName string
imageToAdetailer object
args object[]
Array [
adModel string
The model to use for the adetailer Support list
adPrompt object[]
Array [
text string
weight float
]
adNegativePrompt object[]
Array [
text string
weight float
]
adConfidence float
adDilateErode int32
Default value: 4
adMaskMergeInvert string
Default value: None
adDenoisingStrength float
Default value: 0.4
adInpaintOnlyMasked boolean
Default value: true
adInpaintOnlyMaskedPadding float
Default value: 32
adUseInpaintWidthHeight boolean
Default value: false
adInpaintWidth int32
Default value: 512
adInpaintHeight int32
Default value: 512
adUseSteps boolean
Default value: false
adSteps int32
Default value: 20
adUseCfgScale boolean
Default value: false
adCfgScale float
Default value: 7
lora object
items object[]
Array [
loraModel string
The model to use for the diffusion , How to get the model id
weight float
blockWeight string
lora block weight, value such as <weight>:lbw=<layer weight> example: "1:lbw=1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0"
loraAccessKey string
]
adUseCheckpoint boolean
adCheckpoint string
adUseSampler boolean
adUseNoiseMultiplier boolean
adNoiseMultiplier float
adUseClipSkip boolean
adClipSkip int32
adSampler string
]
diffusion object
width int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
height int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
prompts object[]required
Possible values: <= 150
An array of text prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
negativePrompts object[]required
Possible values: <= 150
An array of text negative prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
sdModel string
The model to use for the diffusion , How to get the model id
sdVae string
The vae to use for the diffusion Support list
sampler string
Which sampler to use for the diffusion process. If this value is omitted we'll automatically select an appropriate sampler for you. Support list
steps int32
Possible values: >= 1 and <= 60
Default value: 0
Number of diffusion steps to run.
cfgScale float
Possible values: <= 30
Default value: 7
How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt)
clipSkip int32
denoisingStrength float
etaNoiseSeedDelta int32
controlnet object
args object[]
Array [
inputImageResourceId string
maskResourceId string
preprocessor stringrequired
The model to use for the controlnet preprocessor Support list
model stringrequired
The model to use for the controlnet Support list
weight float
resizeMode string
Possible values: [DEFAULT, JUST_RESIZE, CROP_AND_RESIZE, RESIZE_AND_FILL]
Default value: DEFAULT
guidance float
guidanceStart float
guidanceEnd float
controlMode string
Possible values: [DEFAULT, BALANCED, MY_PROMPT_IS_MORE_IMPORTANT, CONTROLNET_IS_MORE_IMPORTANT]
Default value: DEFAULT
pixelPerfect boolean
preprocessorParams object
]
lora object
items object[]
Array [
loraModel string
The model to use for the diffusion , How to get the model id
weight float
blockWeight string
lora block weight, value such as <weight>:lbw=<layer weight> example: "1:lbw=1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0"
loraAccessKey string
]
animateDiff object
args object[]
Array [
videoLength int32
fps int32
]
embedding object
items object[]
Array [
model string
The model to use for the negative prompts embedding , How to get the model id
weight float
embeddingAccessKey string
]
v1Clip boolean
modelAccessKey string
scheduleName string
enableElla todo：优化成stages,因为底层还不支持，所以目前优化没用，暂时只在推理部分支持。 (boolean)
enablePix2pix boolean
useHunyuanDit deprecated (boolean)
layerDiffusion object
enable boolean
weight float
clipEncoderName string
imageToInpaint object
resizeMode JUST_RESIZE (string)
Possible values: [DEFAULT, JUST_RESIZE, CROP_AND_RESIZE, RESIZE_AND_FILL, JUST_RESIZE_LATENT_UPSCALE]
Default value: DEFAULT
maskImageResourceId string
maskBlur float
inpaintingFill ORIGINAL (string)
Possible values: [DEFAULT, FILL, ORIGINAL, LATENT_NOISE, LATENT_NOTHING]
Default value: DEFAULT
inpaintFullRes true (boolean)
inpaintFullResPadding int64
inpaintMaskInvert int64
diffusion object
width int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
height int64required
Possible values: >= 512 and <= 1536
Default value: 512
Height of the image in pixels. Must be in increments of 64 and pass the following validation:
• For 512 engines: 262,144 ≤ height * width ≤ 1,048,576, Maximum 1024
• For 768 engines: 589,824 ≤ height * width ≤ 1,048,576, Maximum 1024
• For SDXL v1.0: 262,144 ≤ height * width ≤ 2,073,600, Maximum 1536
prompts object[]required
Possible values: <= 150
An array of text prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
negativePrompts object[]required
Possible values: <= 150
An array of text negative prompts to use for generation. Given a text prompt with the text A lighthouse on a cliff and a weight of 0.5, it would be represented as:
Array [
text string
weight float
]
sdModel string
The model to use for the diffusion , How to get the model id
sdVae string
The vae to use for the diffusion Support list
sampler string
Which sampler to use for the diffusion process. If this value is omitted we'll automatically select an appropriate sampler for you. Support list
steps int32
Possible values: >= 1 and <= 60
Default value: 0
Number of diffusion steps to run.
cfgScale float
Possible values: <= 30
Default value: 7
How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt)
clipSkip int32
denoisingStrength float
etaNoiseSeedDelta int32
controlnet object
args object[]
Array [
inputImageResourceId string
maskResourceId string
preprocessor stringrequired
The model to use for the controlnet preprocessor Support list
model stringrequired
The model to use for the controlnet Support list
weight float
resizeMode string
Possible values: [DEFAULT, JUST_RESIZE, CROP_AND_RESIZE, RESIZE_AND_FILL]
Default value: DEFAULT
guidance float
guidanceStart float
guidanceEnd float
controlMode string
Possible values: [DEFAULT, BALANCED, MY_PROMPT_IS_MORE_IMPORTANT, CONTROLNET_IS_MORE_IMPORTANT]
Default value: DEFAULT
pixelPerfect boolean
preprocessorParams object
]
lora object
items object[]
Array [
loraModel string
The model to use for the diffusion , How to get the model id
weight float
blockWeight string
lora block weight, value such as <weight>:lbw=<layer weight> example: "1:lbw=1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0"
loraAccessKey string
]
animateDiff object
args object[]
Array [
videoLength int32
fps int32
]
embedding object
items object[]
Array [
model string
The model to use for the negative prompts embedding , How to get the model id
weight float
embeddingAccessKey string
]
v1Clip boolean
modelAccessKey string
scheduleName string
enableElla todo：优化成stages,因为底层还不支持，所以目前优化没用，暂时只在推理部分支持。 (boolean)
enablePix2pix boolean
useHunyuanDit deprecated (boolean)
layerDiffusion object
enable boolean
weight float
clipEncoderName string
]
notifyUrl string

Responses

200

OK

application/json

Schema

Schema

job object
id uint64
job id
status string
Possible values: [DEFAULT, CREATED, PENDING, RUNNING, CANCELED, SUCCESS, FAILED, WAITING]
Default value: DEFAULT
job status
credits double
waitingInfo object
waiting info, when status is waiting will return this
queueRank int64
queueLen int64
failedInfo object
failed info, when status is failed will return this
reason string
code string
Possible values: [DEFAULT, TRANSITORY_ERROR, FATAL_ERROR]
Default value: DEFAULT
runningInfo object
running info, when status is running will return this
processingImages object[]
Array [
resourceImage object
id string
url string
expiredIn int64
meta object
image object
format string
width int32
height int32
progress double
]
workflowFinishItem object
status string
Possible values: [DEFAULT, INIT, RUNNING, SUCCESS, FAILED]
Default value: DEFAULT
progress double
step int32
id string
ctime int64
mtime int64
nodes object
property name* object
id string
status string
Possible values: [DEFAULT, INIT, RUNNING, SUCCESS, FAILED]
Default value: DEFAULT
process double
outputUi object
type string
images object[]
Array [
filename string
type string
height int32
width int32
format string
imageResourceId string
]
finishedNodes object
property name* boolean
successInfo object
success info, when status is success will return this
images object[]
final output images
Array [
id string
url string
expiredIn int64
meta object
image object
format string
width int32
height int32
]
videos object[]
final output videos
Array [
id string
url string
expiredIn int64
meta object
image object
format string
width int32
height int32
]
workflowFinishItem object
status string
Possible values: [DEFAULT, INIT, RUNNING, SUCCESS, FAILED]
Default value: DEFAULT
progress double
step int32
id string
ctime int64
mtime int64
nodes object
property name* object
id string
status string
Possible values: [DEFAULT, INIT, RUNNING, SUCCESS, FAILED]
Default value: DEFAULT
process double
outputUi object
type string
images object[]
Array [
filename string
type string
height int32
width int32
format string
imageResourceId string
]
finishedNodes object
property name* boolean

Example (from schema)

{
  "job": {
    "id": "string",
    "status": "DEFAULT",
    "credits": 0,
    "waitingInfo": {
      "queueRank": "string",
      "queueLen": "string"
    },
    "failedInfo": {
      "reason": "string",
      "code": "DEFAULT"
    },
    "runningInfo": {
      "processingImages": [
        {
          "resourceImage": {
            "id": "string",
            "url": "string",
            "expiredIn": "string",
            "meta": {
              "image": {
                "format": "string",
                "width": 0,
                "height": 0
              }
            }
          },
          "progress": 0
        }
      ],
      "workflowFinishItem": {
        "status": "DEFAULT",
        "progress": 0,
        "step": 0,
        "id": "string",
        "ctime": "string",
        "mtime": "string",
        "nodes": {},
        "finishedNodes": {}
      }
    },
    "successInfo": {
      "images": [
        {
          "id": "string",
          "url": "string",
          "expiredIn": "string",
          "meta": {
            "image": {
              "format": "string",
              "width": 0,
              "height": 0
            }
          }
        }
      ],
      "videos": [
        {
          "id": "string",
          "url": "string",
          "expiredIn": "string",
          "meta": {
            "image": {
              "format": "string",
              "width": 0,
              "height": 0
            }
          }
        }
      ],
      "workflowFinishItem": {
        "status": "DEFAULT",
        "progress": 0,
        "step": 0,
        "id": "string",
        "ctime": "string",
        "mtime": "string",
        "nodes": {},
        "finishedNodes": {}
      }
    }
  }
}

400

Bad Request: general error for invalid parameters More specific errors:

• invalid_samples: Sample count may only be greater then 1
• invalid_height_or_width: Height and width must be specified in increments of 64

application/json

Schema

Schema

any

default

Default

application/json

Schema

Schema

code int32
message string
details object[]
Array [
@type string
]

Example (from schema)

{
  "code": 0,
  "message": "string",
  "details": [
    {
      "@type": "string"
    }
  ]
}

Loading...