Platform For AI:Use AI Portrait SDK for Go

Prerequisites

• Go environment is ready.

• 5 to 20 training images and 1 template image are prepared for model training and portrait creation. Supported formats: .jpg, .jpeg, and .png. Each image must be larger than 512×512 pixels.

  • For solo portraits: Template image must contain one face, and all training images must belong to the same person.

  • For group portraits: Template image must contain multiple faces matching the number of model_id values specified during model training.

Preparations

1. Environment dependency: Go 1.18 or later.

2. Install Go SDK locally.

   1. Create a project: go mod init project name

   2. Download the SDK: go get github.com/aliyun/aliyun-pai-aiservice-go-sdk

   3. Import the SDK: import "github.com/aliyun/aliyun-pai-aiservice-go-sdk"

3. Initialize the client.

   1. Set HOST, AppId, and Token as environment variables.

      export HOST="http://ai-service.ce8cc13b6421545749e7b4605f3d02607.cn-hangzhou.alicontainer.com"
      export AppId=xxx
      export Token=xxxxxxxxx

      Parameter	Description
      HOST	Server address: http://ai-service.ce8cc13b6421545749e7b4605f3d02607.cn-hangzhou.alicontainer.com
      AppId	View your AppId on the AI Portrait page after activation.
      Token	View your Token on the AI Portrait page after activation.

   2. Initialize the client.

      host := os.Getenv("Host")
      appId := os.Getenv("AppId")
      token := os.Getenv("Token")
      
      client := sdk.NewClient(host, appId, token)

Sample code

AI Portrait is a resource-intensive service consisting of two main steps: model training and portrait creation. Model training typically takes several minutes, while portrait creation completes in tens of seconds. The following figure shows the API call workflow.

The following sections provide sample request and response code for each API, along with end-to-end examples.

Image check request (client.AIGCApi.AigcImageCheck)

• Sample request code:

  package main
  
  import (
  	"fmt"
  	"log"
  	"os"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
    // Get HOST, AppId, and Token from environment variables.
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	// Check image list.
  	images := []string{"https://train/0.jpg","https://train/1.jpg","https://train/2.jpg","https://train/3.jpg"}
  
  	response, err := client.AIGCApi.AigcImageCheck(images, "", nil)
  
  	fmt.Println(response)
  
  	if err != nil {
  		log.Fatal(err)
  		return
  	}
  
  	if response.Code != "OK" {
  		log.Fatal("response error:%v", response)
  		return
  	}
  
  	// Traverse image check results.
  	for _, result := range response.Data.CheckResults {
  		fmt.Printf("code=%d\tfrontal=%v\turl=%s\n", result.Code, result.Frontal, result.Url)
  	}
  }
  

  Parameters:

  Parameter	Description	Type
  images	Image URLs (comma-separated)	[]string

• Response description:

  The response returns an AIGCImageCheckResponse object containing these fields:

  Parameter	Description	Type
  RequestId	Request ID	string
  Code	Request status code: OK or error	string
  Message	Detailed request status. Returns "success" on success; otherwise returns error details	string
  Data	Returned data details	AIGCImageCheckData

  Data field: An AIGCImageCheckData object.

  Parameter	Description	Type
  CheckResults	Detection result for each input image. Each image corresponds to a dictionary with three keys: url, message, and frontal. These represent image URL, detection details, and whether the image is front-facing	List<AIGCImageCheckResult>
  CostTime	Server-side compute time spent on this API call.	float64
  Images	List of verified image URLs.	List<String>
  RequestId	Request ID (same as the top-level request_id).	String

  check_results field: An AIGCImageCheckResult object.

  Parameter	Description	Type
  Code	Request status code indicating completion: 0 or 1. Detection result for each input image.	int
  Frontal	Image detection details and whether it is front-facing.	bool
  Message	Status information.	string
  Url	Image URL.	string

  Summary of check_results messages:

  message	Status code	Description
  success	1	The requirements are met.
  Image decode error.	2	The image cannot be downloaded or decoded.
  Number of face is not 1.	3	The number of faces in the image is not 1.
  Image detect error.	4	Face detection failed.
  Image encoding error.	5	Failed to encode the face into a feature vector. This error usually occurs if no face is detected.
  This photo is not the same person in photos.	6	The face in this image does not match the faces in the other images.

Model training initiation (client.AIGCApi.AigcImagesTrain)

• Stable Diffusion 1.5 training

  package main
  
  import (
  	"fmt"
  	"log"
  	"os"
  	"time"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
    // Get HOST, AppId, and Token from environment variables.
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	images := []string{"https://train/0.jpg","https://train/1.jpg","https://train/2.jpg","https://train/3.jpg"}
  
  	response, err := client.AIGCApi.AigcImagesTrain(images, "", nil)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	// Job ID
  	jobId := response.Data.JobId
  	modelId := response.Data.ModelId
  	fmt.Println(jobId)
  
  	jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  	fmt.Println(jobResponse.Data.Job.Message)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	for {
  		jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  		fmt.Println(jobResponse.Data.Job.Result)
  		if err != nil {
  			log.Fatal("get job fail", err)
  		}
  
  		if jobResponse.Data.Job.State == sdk.JOB_STATE_WAIT {
  			fmt.Println("job running")
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_SUCCESS { // job success
  			fmt.Println(jobResponse)
  			fmt.Println("job success")
  			break
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_FAILED {
  			log.Fatal("job fail", err)
  			return
  		}
  
  		time.Sleep(30 * time.Second)
  	}
  	fmt.Println(modelId)
  
  }
  

  Parameter Name	Description	Type
  images	Training image URLs	[]string

• Stable Diffusion XL (SDXL) training

  To use SDXL, contact your account manager to activate the service. Then, specify the model name.

  SDXL supports both portrait training and scene LoRA training.

  The training for both is specified by the configuration parameters.

  package main
  
  import (
  	"fmt"
  	"log"
  	"os"
  	"time"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
    // Get HOST, AppId, and Token from environment variables.
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	images := []string{"https://train/0.jpg","https://train/1.jpg","https://train/2.jpg","https://train/3.jpg"}
  
  	modelName := "train_xl"
  	configure := make(map[string]interface{}, 10)
  	configure["train_scene_lora_bool"] = false
  	configure["scene_lora_prompts"] = []string{"a photography of a woman with long blonde hair and a white dress",
  		"a photography of a woman in a pink dress posing for a picture",
  		"a photography of a woman in a black dress with a white background",
  		"a photography of a woman with a frilly collar and suspenders",
  		"a photography of a woman with a white dress and a white headpiece"}
  
  	response, err := client.AIGCApi.AigcImagesTrain(images, modelName, configure)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	// Job ID
  	jobId := response.Data.JobId
  	modelId := response.Data.ModelId
  	fmt.Println(jobId)
  
  	jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  	fmt.Println(jobResponse.Data.Job.Message)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	for {
  		jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  		fmt.Println(jobResponse.Data.Job.Result)
  		if err != nil {
  			log.Fatal("get job fail", err)
  		}
  
  		if jobResponse.Data.Job.State == sdk.JOB_STATE_WAIT {
  			fmt.Println("job running")
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_SUCCESS { // job success
  			fmt.Println(jobResponse)
  			fmt.Println("job success")
  			break
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_FAILED {
  			log.Fatal("job fail", err)
  			return
  		}
  
  		time.Sleep(30 * time.Second)
  	}
  	fmt.Println(modelId)
  
  }
  

  Parameter Name	Description	Type
  images	List of training image URLs.	[]string
  modelName	Model name. Set to train_xl for SDXL.	string
  configure	Parameter configuration (default: nil)	map[string]interface{}

  Parameter in configure	Description	Type	Required	Default	Valid values
  train_scene_lora_bool	Whether to train scene LoRA.	bool	No	False	True, False
  scene_lora_prompts	Prompts for scene LoRA training. Required for scene LoRA training. List length must match the number of image URLs.	[]string	No	[]	N/A

• Response parsing

  The service returns an AIGCImageTrainResponse object.

  Parameter	Description	Type
  RequestId	Request ID.	string
  Code	Request status code indicating completion: OK or error.	string
  Message	Detailed request status. Returns success on success; otherwise, returns specific details.	string
  Data	Details of returned data.	AIGCImageTrainData

  Fields of the AIGCImageTrainData object

  Parameter Name	Description	Type
  jobId	Job ID.	int32
  modelId	Model ID for this training session, a 36-character string.	string

  Save the jobId to query the training results.

  Save the modelId to request portrait creation.

Training result query (client.JobApi.GetAsyncJobWithId)

• Sample request code

  package main
  
  import (
  	"fmt"
  	"log"
  	"os"
  	"time"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
  	// Get HOST, AppId, and Token from environment variables.
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	// Job ID
  	var jobId int32 = xxxxxx
  
  	for {
  		jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  		if err != nil {
  			log.Fatal("get job fail", err)
  		}
  
  		if jobResponse.Data.Job.State == sdk.JOB_STATE_WAIT {
  			fmt.Println("job running")
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_SUCCESS { // job success
  			fmt.Println(jobResponse)
  			fmt.Println("job success")
  			break
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_FAILED {
  			log.Fatal("job fail", err)
  			return
  		}
  
  		time.Sleep(30 * time.Second)
  	}
  
  }
  

  Parameters:

  Parameter	Description	Type	Required
  jobId	Job ID returned in the service response.	int32	Yes

• The following describes how the response is parsed.

  The service returns an AsyncJobResponse object with the following fields:

  Parameter	Description	Type
  RequestId	Request ID.	string
  Code	Request status code indicating completion: OK or error.	string
  Message	Detailed request status. Returns success on success; otherwise, returns specific details.	string
  Data	Details of returned data.	AsyncJobData

  Parsing the AsyncJobData.Job field (AsyncJobResult)

  Parameter	Description	Type
  State	Verification result for each image.	int32
  AppId	Account AppId.	string
  Message	Detailed request status. Returns success on success; otherwise, returns specific details.	string
  Result	Model output.	string

  After model training is complete, the Result field contains the following:

  Parameter	Description
  cost_time	Total training time consumed.
  states	Verification result for each image. Detection result for each input image. Each image corresponds to a dictionary with three keys: url, message, and frontal. These represent the image URL, detection details, and whether the image is front-facing.
  model_id	LoRA model name, identical to model_id from training LoRA model name from training, used as input for portrait creation

Portrait creation request

• Solo portrait creation (client.AIGCApi.AigcImagesCreate)

  • Stable Diffusion 1.5 prediction

    package main
    
    import (
    	"encoding/base64"
    	"log"
    	"os"
    
    	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
    )
    
    func main() {
      // Get HOST, AppId, and Token from environment variables.
    	host := os.Getenv("Host")
    	appId := os.Getenv("AppId")
    	token := os.Getenv("Token")
    
    	client := sdk.NewClient(host, appId, token)
    
    	modelID := "xxxxxxxxxxxxxxx"
    
    	template_image := "https://template_case.png"
    
    	config := make(map[string]interface{}, 10)
    	config["lora_weights"] = 0.9
    	config["first_denoising_strength"] = 1
    	config["second_denoising_strength"] = 1
    	config["more_like_me"] = 1
    	config["crop_face_preprocess"] = false
    	config["apply_face_fusion_before"] = false
    	config["apply_face_fusion_after"] = false
    	config["color_shift_middle"] = false
    	config["color_shift_last"] = false
    	config["background_restore"] = true
    
    	response, err := client.AIGCApi.AigcImagesCreate(modelID, image, "", config)
    
    	if err != nil {
    		log.Fatal(err)
    	}
    
    	image_decode, _ := base64.StdEncoding.DecodeString(response.Data.Image)
    
    	f, err_2 := os.Create("test.jpg")
    	f.Write(image_decode)
    
    	if err_2 != nil {
    		log.Fatal(err)
    	}
    
    }
    

    Parameter	Description	Type	Required
    modelID	LoRA model name. Use the model-id obtained from training.	string	Yes
    template_image	Template image URL	string	Yes
    config	Model return configuration (default: nil)	map[string]interface{}	Yes

    Parameters in config

    Internal Parameter Name	Description	Type	Required	Default	Valid values
    lora_weights	LoRA strength.	float	No	0.90	0.5–1.0
    first_denoising_strength	Strength of the first image reconstruction.	float	No	0.45	0.0–1.0
    second_denoising_strength	Strength of the second image reconstruction.	float	No	0.30	0.0–1.0
    more_like_me	That's more my speed.	float	No	0.50	0.0–1.0
    crop_face_preprocess	Whether to crop the image before reconstruction. Recommended for large images.	bool	No	True	True, False
    apply_face_fusion_before	Whether to perform the first portrait merging.	bool	No	True	True, False
    apply_face_fusion_after	Whether to perform the second portrait merging.	bool	No	True	True, False
    color_shift_middle	Whether to perform the first color correction.	bool	No	True	True, False
    color_shift_last	Whether to perform the second color correction.	bool	No	True	True, False
    background_restore	Whether to reconstruct the background.	bool	No	False	True, False
    skin_retouching_bool	Whether to perform skin smoothing. When enabled, smooths and brightens skin, generally improving appearance but potentially causing overly pale skin. Disabling improves skin texture.	bool	No	False	True, False
    photo_enhancement_bool	Whether to enhance the portrait. When enabled, performs portrait restoration or super resolution to improve image quality.	bool	No	True	True, False
    photo_enhancement_method	Portrait enhancement method. photo_fix performs image restoration, which may cause some distortion but fixes unreasonable areas and may reduce skin texture. super_resolution only performs super resolution, preserving more of the original image.	string	No	photo_fix	photo_fix, super_resolution
    makeup_transfer	Whether to perform makeup transfer. When enabled, transfers makeup to prevent overly plain images, but may reduce resemblance to the user.	bool	No	False	True, False
    makeup_transfer_ratio	Strength of makeup transfer. Higher values increase makeup transfer proportion and resemblance to the template, but may reduce user resemblance.	float	No	0.5	0.0–1.0
    face_shape_match	Whether to adapt face shape. When enabled, reduces control strength, making face shape and skin texture closer to the user.	bool	No	False	True, False
    ipa_control	Whether to enable IP-Adapter control. When enabled, increases portrait similarity but makes results more sensitive to reference images.	bool	No	False	True, False
    ipa_control_only	When enabled, allows prediction without training. Requires ipa_image_path when enabled.	bool	No	False	True, False
    ipa_image_path	Reference portrait URL. Required when ipa_control_only is enabled.	string	No	None	Downloadable URL
    ipa_weight	IP-Adapter control strength. Higher values increase user resemblance but may cause image distortion if too high.	float	No	0.5	0.0–1.0
    style_name	Set the style for the generated output Realistic or Anime style.	string	No	Realistic	Realistic, Anime
    lcm_accelerate	Controls the generation style and the use of LCM acceleration.	bool	No	False	True, False
    sharp_ratio	Sharpening level. Appropriate values improve clarity. Excessive values cause image distortion.	float	No	0.15	0.0–1.0
    t2i_prompt	Set this parameter to use text-to-image generation. If t2i_prompt is set, you do not need to provide a template URL.	String	No	None	N/A

  • SDXL prediction

    To use SDXL, contact your account manager to activate the service. Then, specify the model name.

    package main
    
    import (
    	"encoding/base64"
    	"log"
    	"os"
    
    	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
    )
    
    func main() {
      // Get HOST, AppId, and Token from environment variables.
    	host := os.Getenv("Host")
    	appId := os.Getenv("AppId")
    	token := os.Getenv("Token")
    
    	client := sdk.NewClient(host, appId, token)
    
    	modelID := "xxxxxxxxxxxx"
    
    	template_image := "https://template_case.png"
    
    	modelName := "create_xl"
    	config := make(map[string]interface{}, 10)
    	// config["lora_weights"] = 0.9
    	// config["first_denoising_strength"] = 1
    	// config["second_denoising_strength"] = 1
    	// config["more_like_me"] = 1
    	// config["crop_face_preprocess"] = false
    	// config["apply_face_fusion_before"] = false
    	// config["apply_face_fusion_after"] = false
    	// config["color_shift_middle"] = false
    	// config["color_shift_last"] = false
    	// config["background_restore"] = true
    
    	response, err := client.AIGCApi.AigcImagesCreate(modelID, image, modelName, config)
    
    	if err != nil {
    		log.Fatal(err)
    	}
    
    	image_decode, _ := base64.StdEncoding.DecodeString(response.Data.Image)
    
    	f, err_2 := os.Create("test.jpg")
    	f.Write(image_decode)
    
    	if err_2 != nil {
    		log.Fatal(err)
    	}
    
    }
    

    Parameter Name	Type	Description
    modelID	string	LoRA model name from training. Set to "" when using ipa_control_only mode
    template_image	string	The URL path for the template. Set to "t2i_generate" when using scene_lora or prompt generation.
    modelName	string	Model name. Set to create_xl for SDXL.
    config	map[string]interface{}	Model return configuration (default: None)

    Parameters in config

    Internal configuration parameter name	Description	Type	Required	Default	Valid values
    lora_weights	LoRA application strength. Higher values increase user resemblance but may cause image distortion if too high.	float	No	0.90	0.5–1.0
    first_diffusion_steps	Steps for first diffusion. Changing this value is not recommended. A value that is too low will cause image aliasing.	int	No	50	20–50
    first_denoising_strength	Strength of the first image reconstruction. Face reconstruction strength. Higher values increase reconstruction and user resemblance, but excessive values cause image inconsistency.	float	No	0.45	0.0–1.0
    second_diffusion_steps	Steps for second diffusion. Do not modify. Too few steps cause image distortion.	int	No	30	20–50
    second_denoising_strength	Strength of the second image reconstruction. Facial edge reconstruction strength. Excessive values cause image inconsistency.	float	No	0.30	0.0–1.0
    more_like_me	That's more my speed. Portrait merging ratio. Higher values increase resemblance to the subject, but excessive values reduce realism.	float	No	0.60	0.0–1.0
    crop_face_preprocess	Whether to crop the image before reconstruction. Recommended for large images. Do not adjust.	bool	No	True	True, False
    apply_face_fusion_before	Whether to perform the first portrait merging. When enabled, performs portrait merging. Disabling reduces similarity.	bool	No	True	True, False
    apply_face_fusion_after	Whether to perform the second portrait merging. When enabled, performs portrait merging. Disabling reduces similarity.	bool	No	True	True, False
    color_shift_middle	Whether to perform the first color correction. When enabled, corrects colors to match template skin tone. Disabling may cause color bias.	bool	No	True	True, False
    color_shift_last	Whether to perform the second color correction. When enabled, corrects colors to match template skin tone. Disabling may cause color bias.	bool	No	True	True, False
    background_restore	Whether to reconstruct the background. When enabled, reconstructs background for a more natural look, but changes the background and increases processing time.	bool	No	False	True, False
    skin_retouching_bool	Whether to perform skin smoothing. When enabled, smooths and brightens skin, generally improving appearance but potentially causing overly pale skin. Disabling improves skin texture.	bool	No	False	True, False
    photo_enhancement_bool	Whether to enhance the portrait. When enabled, performs portrait restoration or super resolution to improve image quality.	bool	No	True	True, False
    photo_enhancement_method	Portrait enhancement method: photo_fix performs image restoration, which may cause some distortion but fixes unreasonable areas and may reduce skin texture. super_resolution only performs super resolution, preserving more of the original image.	String	No	photo_fix	photo_fix, super_resolution
    makeup_transfer	Whether to perform makeup transfer. When enabled, transfers makeup to prevent overly plain images, but may reduce user resemblance.	bool	No	False	True, False
    makeup_transfer_ratio	Strength of makeup transfer. Higher values increase makeup transfer proportion and resemblance to the template, but may reduce user resemblance.	float	No	0.5	0.0–1.0
    ipa_control	Whether to enable IP-Adapter control. When enabled, increases portrait similarity but makes results more sensitive to reference images.	bool	No	False	True, False
    ipa_control_only	When enabled, allows prediction without training. Requires ipa_image_path when enabled.	bool	No	False	True, False
    ipa_image_path	Reference portrait URL. Required when ipa_control_only is enabled.	String	No	None	Downloadable URL
    ipa_weight	IP-Adapter control strength. Higher values increase user resemblance but may cause image distortion if too high.	float	No	0.5	0.0–1.0
    lcm_accelerate	Specifies what generation style to use and whether to use LCM acceleration.	bool	No	False	True, False
    sharp_ratio	Sharpening level. Appropriate values improve clarity. Excessive values cause image distortion.	float	No	0.15	0.0–1.0
    scene_id	Set this parameter to use scene lora. scene_id comes from training. The model_id returned during SDXL scene LoRA training is the scene_id.	String	No	None	N/A
    t2i_prompt	Set this parameter to use text-to-image generation. If t2i_prompt is set, you do not need to provide a template URL.	String	No	None	N/A

• Group portrait creation (client.AIGCApi.AigcImagesCreateByMultiModelIds)

  package main
  
  import (
  	"encoding/base64"
  	"log"
  	"os"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
    // Get HOST, AppId, and Token from environment variables.
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	modelIDs := []string{"xxxxxx", "xxxxxx"}
  
  	template_image := "https://template_case.png"
  
  	config := make(map[string]interface{}, 10)
  	config["lora_weights"] = 0.9
  	config["first_denoising_strength"] = 1
  	config["second_denoising_strength"] = 1
  	config["more_like_me"] = 1
  	config["crop_face_preprocess"] = false
  	config["apply_face_fusion_before"] = false
  	config["apply_face_fusion_after"] = false
  	config["color_shift_middle"] = false
  	config["color_shift_last"] = false
  	config["background_restore"] = true
  
  	response, err := client.AIGCApi.AigcImagesCreateByMultiModelIds(modelIDs, image, "", config)
  
  	image_decode, _ := base64.StdEncoding.DecodeString(response.Data.Image)
  
  	f, err := os.Create("test.jpg")
  	f.Write(image_decode)
  
  	if err != nil {
  		log.Fatal(err)
  	}
  
  }
  

  Parameter Name	Description	Type	Required
  modelIds	model_id values of all LoRA models used for image generation.	[]string	Yes
  template_image	The path to the template.	string	Yes
  config	Model return configuration (default: nil)	map[string]interface{}	Yes

• Response parsing:

  The service returns an AIGCImageCreateResponse object with the following fields:

  Parameter Name	Description	Type
  RequestId	Request ID.	string
  Code	Request status code indicating completion: OK or error.	string
  Message	Detailed request status. Returns success on success; otherwise, returns specific details.	string
  Data	Details of returned data.	AIGCImageCreateData

  Fields of the AIGCImageCreateData data type:

  Parameter	Description	Type
  CostTime	Time spent generating the image.	Float
  Image	Base64-encoded image.	String

• Error code descriptions

  1. Service request error codes:

     HTTP status code	code	message	Description
     400	PARAMETER_ERROR	not found appid	Invalid AppId.
     400	EXCEEDED_QUOTA_ERROR	exceeded quota	Account call quota exhausted.
     401	PARAMETER_ERROR	sign error	Invalid token.
     404	PARAMETER_ERROR	model not found	Requested model service not deployed.

  2. Result query error codes:

     HTTP status code	code	message	Description
     462	error	Invalid input data. Please check the input dict.	Input data parsing error.
     462	error	Image not provided. Please check the template_image.	No template image provided for portrait creation.
     462	error	Prompts get error. Please check the model_id.	Check the format of the provided model_id.
     462	error	Roop image decode error. Please check the user's lora is trained or not.	Roop image does not exist. Check if the model is trained.
     462	error	Template image decode error. Provide a new template	Template image decode error. Provide a new template
     462	error	There is not face in template. Please Give a new template.	No face found in template image. Provide a new template.
     462	error	Template image process error. Please Give a new template.	Template image preprocessing error. Provide a new template.
     469	error	First Face Fusion Error, Can't get face in template image.	First portrait merging failed.
     469	error	First Stable Diffusion Process error. Check the webui status.	First Stable Diffusion processing failed.
     469	error	Second Face Fusion Error, Can't get face in template image.	Second portrait merging failed.
     469	error	Second Stable Diffusion Process error. Check the webui status.	Second Stable Diffusion processing failed.
     469	error	Please confirm if the number of faces in the template corresponds to the user ID.	Verify that the number of user IDs matches the number of faces.
     469	error	Third Stable Diffusion Process error. Check the webui status.	Background processing failed. Replace the template.
     500	error	Face id image decode error. Pleace check the user's lora is trained or not.	Face ID image decoding error. Check if the model is trained.

End-to-end workflow sample code

• Standard workflow (SD15)

  package main
  
  import (
  	"encoding/base64"
  	"fmt"
  	"log"
  	"os"
  	"time"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  	images := []string{"https://train/0.jpg","https://train/1.jpg","https://train/2.jpg","https://train/3.jpg"}
  		
  	trainresponse, err := client.AIGCApi.AigcImagesTrain(images, "", nil)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	jobId := trainresponse.Data.JobId
  	modelId := trainresponse.Data.ModelId
  	fmt.Println(jobId)
  
  	jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  	fmt.Println(jobResponse.Data.Job.Message)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	for {
  		jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  		fmt.Println(jobResponse.Data.Job.Result)
  		if err != nil {
  			log.Fatal("get job fail", err)
  		}
  
  		if jobResponse.Data.Job.State == sdk.JOB_STATE_WAIT {
  			fmt.Println("job running")
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_SUCCESS { // job success
  			fmt.Println(jobResponse)
  			fmt.Println("job success")
  			break
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_FAILED {
  			log.Fatal("job fail", err)
  			return
  		}
  
  		time.Sleep(30 * time.Second)
  
  	}
  	fmt.Println(modelId)
  
  	template_image := "https://template_case.png"
  
  	config := make(map[string]interface{}, 10)
  	config["lora_weights"] = 0.9
  	config["first_denoising_strength"] = 1
  	config["second_denoising_strength"] = 1
  	config["more_like_me"] = 1
  	config["crop_face_preprocess"] = false
  	config["apply_face_fusion_before"] = false
  	config["apply_face_fusion_after"] = false
  	config["color_shift_middle"] = false
  	config["color_shift_last"] = false
  	config["background_restore"] = true
  
  	createresponse, err := client.AIGCApi.AigcImagesCreate(modelId, template_image, "", config)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	image_decode, _ := base64.StdEncoding.DecodeString(createresponse.Data.Image)
  
  	f, err := os.Create("test.jpg")
  	f.Write(image_decode)
  
  	if err != nil {
  		log.Fatal(err)
  	}
  }
  

  Parameter	Description	Type	Required
  images	List of training image URLs.	[]string	Yes
  template_image	The template's URL path.	string	Yes
  config	Model return configuration. Default: nil.	map[string]interface{}	Yes

• Standard workflow (SDXL)

  package main
  
  import (
  	"encoding/base64"
  	"fmt"
  	"log"
  	"os"
  	"time"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  	images := []string{
              "https://xxx/0.jpg",
              "https://xxx/1.jpg",
              "https://xxx/2.jpg"}
                       
  	trainresponse, err := client.AIGCApi.AigcImagesTrain(images, "train_xl", nil)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	jobId := trainresponse.Data.JobId
  	modelId := trainresponse.Data.ModelId
  	fmt.Println(jobId)
  
  	jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  	fmt.Println(jobResponse.Data.Job.Message)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	for {
  		jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  		fmt.Println(jobResponse.Data.Job.Result)
  		if err != nil {
  			log.Fatal("get job fail", err)
  		}
  
  		if jobResponse.Data.Job.State == sdk.JOB_STATE_WAIT {
  			fmt.Println("job running")
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_SUCCESS { // job success
  			fmt.Println(jobResponse)
  			fmt.Println("job success")
  			break
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_FAILED {
  			log.Fatal("job fail", err)
  			return
  		}
  
  		time.Sleep(30 * time.Second)
  
  	}
  	fmt.Println(modelId)
  
  	template_image = "https://demo.jpg"
  
  	config := make(map[string]interface{}, 10)
  
  	createresponse, err := client.AIGCApi.AigcImagesCreate(modelId, template_image, "create_xl", config)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	image_decode, _ := base64.StdEncoding.DecodeString(createresponse.Data.Image)
  
  	f, err := os.Create("test.jpg")
  	f.Write(image_decode)
  
  	if err != nil {
  		log.Fatal(err)
  	}
  }
  

  Parameter Name	Description	Type	Required
  images	List of training image URLs.	[]string	Yes
  template_image	Template URL path	string	Yes
  config	Model return configuration. Default: nil.	map[string]interface{}	Yes

• Creating an ingest endpoint from a single reference graph without training

  package main
  
  import (
  	"encoding/base64"
  	"log"
  	"os"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	template_image = "https://demo.jpg"
    ref_image = "https://reference.jpg"
    
  	config := make(map[string]interface{}, 10)
  	config["ipa_control_only"] = true
  	config["ipa_weight"] = 0.6
  	config["ipa_image_path"] = ref_image
  
  	createresponse, err := client.AIGCApi.AigcImagesCreate("", template_image, "", config)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	image_decode, _ := base64.StdEncoding.DecodeString(createresponse.Data.Image)
  
  	f, err := os.Create("test.jpg")
  	f.Write(image_decode)
  
  	if err != nil {
  		log.Fatal(err)
  	}
  }
  

  Parameter name	Description	Type	Required
  template_image	The template's URL path.	string	Yes
  ref_image	Reference image URL	string	Yes
  config	Model return configuration. Default: nil.	map[string]interface{}	Yes

• Generate a template creation workflow from prompts

  package main
  
  import (
  	"encoding/base64"
  	"fmt"
  	"log"
  	"os"
  	"time"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  	images := []string{"https://xxx/0.jpg",
        							 "https://xxx/1.jpg",
                       "https://xxx/2.jpg"}
  	trainresponse, err := client.AIGCApi.AigcImagesTrain(images, "", nil)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	jobId := trainresponse.Data.JobId
  	modelId := trainresponse.Data.ModelId
  	fmt.Println(jobId)
  
  	jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  	fmt.Println(jobResponse.Data.Job.Message)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	for {
  		jobResponse, err := client.JobApi.GetAsyncJobWithId(jobId)
  		fmt.Println(jobResponse.Data.Job.Result)
  		if err != nil {
  			log.Fatal("get job fail", err)
  		}
  
  		if jobResponse.Data.Job.State == sdk.JOB_STATE_WAIT {
  			fmt.Println("job running")
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_SUCCESS { // job success
  			fmt.Println(jobResponse)
  			fmt.Println("job success")
  			break
  		} else if jobResponse.Data.Job.State == sdk.JOB_STATE_FAILED {
  			log.Fatal("job fail", err)
  			return
  		}
  
  		time.Sleep(30 * time.Second)
  
  	}
  	fmt.Println(modelId)
  
  	template_image = "https://demo.jpg"
  
  	t2i_prompt := "(portrait:1.5), 1girl, bokeh, bouquet, brown_hair, cloud, flower, hairband, hydrangea, lips, long_hair, outdoors, sunlight, white_flower, white_rose, green sweater, sweater, (cloth:1.0), (best quality), (realistic, photo-realistic:1.3), film photography, minor acne, (portrait:1.1), (indirect lighting), extremely detailed CG unity 8k wallpaper, enormous filesize, best quality, realistic, photo-realistic, ultra high res, raw photo, put on makeup"
  
  	config := make(map[string]interface{}, 10)
  	config["t2i_prompt"] = t2i_prompt
  
  	createresponse, err := client.AIGCApi.AigcImagesCreate(modelId, template_image, "", config)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	image_decode, _ := base64.StdEncoding.DecodeString(createresponse.Data.Image)
  
  	f, err := os.Create("test.jpg")
  	f.Write(image_decode)
  
  	if err != nil {
  		log.Fatal(err)
  	}
  }
  

  Parameter Name	Description	Type	Required
  images	List of images for LoRA training.	[]string	Yes
  template_image	The URL path for the template.	string	Yes
  t2i_prompt	Prompt	string	Yes
  config	Model return configuration. Default: nil.	map[string]interface{}	Yes

• Generate a template using a prompt and a single reference image without a template or training

  package main
  
  import (
  	"encoding/base64"
  	"log"
  	"os"
  
  	sdk "github.com/aliyun/aliyun-pai-aiservice-go-sdk"
  )
  
  func main() {
  	host := os.Getenv("Host")
  	appId := os.Getenv("AppId")
  	token := os.Getenv("Token")
  
  	client := sdk.NewClient(host, appId, token)
  
  	template_image = "https://demo.jpg"
    ref_image = "https://reference.jpg"
    
  	t2i_prompt := "(portrait:1.5), 1girl, bokeh, bouquet, brown_hair, cloud, flower, hairband, hydrangea, lips, long_hair, outdoors, sunlight, white_flower, white_rose, green sweater, sweater, (cloth:1.0), (best quality), (realistic, photo-realistic:1.3), film photography, minor acne, (portrait:1.1), (indirect lighting), extremely detailed CG unity 8k wallpaper, enormous filesize, best quality, realistic, photo-realistic, ultra high res, raw photo, put on makeup"
  
  	config := make(map[string]interface{}, 10)
  	config["ipa_control_only"] = true
  	config["ipa_weight"] = 0.6
  	config["ipa_image_path"] = ref_image
  	config["t2i_prompt"] = t2i_prompt
  
  	createresponse, err := client.AIGCApi.AigcImagesCreate("", template_image, "", config)
  	if err != nil {
  		log.Fatal(err)
  	}
  
  	image_decode, _ := base64.StdEncoding.DecodeString(createresponse.Data.Image)
  
  	f, err := os.Create("test.jpg")
  	f.Write(image_decode)
  
  	if err != nil {
  		log.Fatal(err)
  	}
  }
  

  Parameter	Description	Type	Required
  ref_image	Reference image URL path.	string	Yes
  template_image	Template path	string	Yes
  t2i_prompt	Prompt.	string	Yes
  config	Model return configuration. Default: nil.	map[string]interface{}	Yes

References

For more information, see the Go SDK GitHub repository.