Get deployment details

Prerequisites

Supported region: The features described in this document are only available in the China (Beijing) region, and you must use the API Key from that region.
You have successfully obtained an API Key and configured it as an environment variable.
You have read Model deployment and Deploy models using the API to understand the basic steps of model deployment.

Get model deployment status

Query the details and running status of a specified model deployment. You can poll this API; when the status becomes RUNNING, it indicates that the model has been deployed successfully.

Note

Model deployment is expected to take 5 to 10 minutes.

Endpoint

GET https://dashscope.aliyuncs.com/api/v1/deployments/{deployed_model}

Request example

Sample request:

curl "https://dashscope.aliyuncs.com/api/v1/deployments/qwen-plus-202305099980-fac9-sample" \
    --header "Authorization: Bearer ${DASHSCOPE_API_KEY}" \
    --header 'Content-Type: application/json'

Request parameters

Parameter	Type	Location	Required	Description
deployed_model	String	path	Yes	The unique identifier for the model deployment. Returned by Create deployment or List deployments.

Response example

Pay attention to the status field. When the status becomes RUNNING, the model has been deployed successfully and is ready for invocation.

{
  "request_id": "66a855f0-a6fe-4b05-9786-fb30c7c6782d",
  "output": {
    "deployed_model": "emo-35b3f106-sample01",
    "gmt_create": "2025-06-17T11:00:38",
    "gmt_modified": "2025-06-17T11:06:13",
    "status": "RUNNING",
    "model_name": "emo",
    "base_model": "emo",
    "base_capacity": 1,
    "capacity": 1,
    "ready_capacity": 1,
    "workspace_id": "llm-v71tlv3***",
    "charge_type": "post_paid",
    "creator": "175805416***",
    "modifier": "175805416***"
  }
}

Response parameters

Field	Type	Description
request_id	String	Unique identifier for the request.
output	Object	Job details.
output.deployed_model	String	Unique identifier of the model deployment. Used to query deployment status and invoke the model.
output.model_name	String	Model identifier name.
output.base_model	String	The base model used.
output.status	String	Deployment status: PENDING: The deployment is being created. RUNNING: The deployment is running and the model can process requests normally. UPDATING: Configuration change in progress. UPDATING_FAILED: Configuration change failed. DELETING: The deployment is being deleted. DELETED: The deployment has been deleted. FAILED: Deployment creation or update failed. STOPPED: The deployment has been stopped and will not be charged. ARREARS_DOWN: Service suspended due to overdue payment. ARREARS_RECOVERING: Recovering from payment suspension. OFFLINING: Service is being taken offline.
output.base_capacity	Number	Base resource unit count.
output.capacity	Number	Current resource unit count.
output.ready_capacity	Number	Number of ready resource units.
output.workspace_id	String	The workspace ID associated with the Alibaba Cloud Model Studio API Key. See Get the Workspace ID.
output.charge_type	String	Billing mode. post_paid indicates pay-as-you-go.
output.gmt_create	String	Deployment creation time.
output.gmt_modified	String	Deployment last modified time.
output.creator	String	Alibaba Cloud account ID of the creator.
output.modifier	String	Alibaba Cloud account ID of the modifier.
output.plan	String	Deployment plan.

List deployable models

Get the list of eligible candidate models for deployment on the Model Studio platform, i.e., which models can be used to create deployments.

Endpoint

GET https://dashscope.aliyuncs.com/api/v1/deployments/models

Request example

Use the following command to query models available for deployment. We recommend using version=v1.0 to get a complete response including deployment plans and template information.

curl "https://dashscope.aliyuncs.com/api/v1/deployments/models?page_no=1&page_size=100&version=v1.0&model_source=base" \
    --header "Authorization: Bearer $DASHSCOPE_API_KEY" \
    --header 'Content-Type: application/json'

Query user fine-tuned models:

curl "https://dashscope.aliyuncs.com/api/v1/deployments/models?page_no=1&page_size=100&version=v1.0&model_source=custom" \
    --header "Authorization: Bearer $DASHSCOPE_API_KEY" \
    --header 'Content-Type: application/json'

Request parameters

Parameter	Type	Required	Description
page_no	Number	No	Page number. Default value: 1.
page_size	Number	No	Page size. Default value: 50. Maximum value: 100. Minimum value: 1.
model_source	String	No	Model source. `base` indicates system models (default), `custom` indicates user fine-tuned models.
version	String	No	API version. We recommend using `v1.0`. When `v1.0` is specified, the response will include complete deployment plan and template information.

Response example

After the command is executed, the following result is returned:

{
    "request_id": "f7da015c-ea90-4d96-af89-2f8d7604026a",
    "output": {
        "page_no": 1,
        "page_size": 100,
        "total": 5,
        "models": [
            {
                "model_name": "qwen3-8b",
                "plans": [
                    {
                        "plan": "mu",
                        "templates": [
                            {
                                "template_id": "MU1",
                                "template_name": "Single-node deployment - Standard inference",
                                "template_type": "COUPLED",
                                "template_version": "v1",
                                "template_desc": "Suitable for standard inference scenarios",
                                "roles": {
                                    "unified": {
                                        "model_unit_spec": "MU1",
                                        "capacity_unit_per_instance": 4
                                    }
                                }
                            },
                            {
                                "template_id": "MU1-PD",
                                "template_name": "PD-separated deployment - Standard inference",
                                "template_type": "SEPERATED",
                                "template_version": "v1",
                                "template_desc": "Suitable for PD-separated inference scenarios",
                                "roles": {
                                    "prefill": {
                                        "model_unit_spec": "MU1",
                                        "capacity_unit_per_instance": 4
                                    },
                                    "decode": {
                                        "model_unit_spec": "MU1",
                                        "capacity_unit_per_instance": 4
                                    }
                                }
                            }
                        ]
                    },
                    {
                        "plan": "lora"
                    }
                ]
            }
        ]
    }
}

Response parameters

Parameter	Type	Description
models	Array	List of deployable models.
models[].model_name	String	Model name.
models[].plans	Array	List of deployment plans supported by the model. Returned when `version=v1.0` is specified.
models[].plans[].plan	String	Deployment plan type: `mu` (Model Unit), `cu` (Compute Unit), `ptu` (Provisioned Throughput Unit), `lora` (LoRA shared deployment).
models[].plans[].templates	Array	List of deployment templates (returned when `plan=mu`).
models[].plans[].templates[].template_id	String	Template ID, passed as the `deploy_spec` parameter when creating a deployment.
models[].plans[].templates[].template_name	String	Template display name.
models[].plans[].templates[].template_type	String	Template type: `COUPLED` (non PD-separated, uses the `capacity` parameter), `SEPERATED` (PD-separated, uses the `prefill_capacity` and `decode_capacity` parameters).
models[].plans[].templates[].template_version	String	Template version.
models[].plans[].templates[].template_desc	String	Template description.
models[].plans[].templates[].roles	Object	Node role configuration. COUPLED mode contains a `unified` node; SEPERATED mode contains `prefill` and `decode` nodes.
models[].plans[].templates[].roles.{role}.model_unit_spec	String	Model unit specification.
models[].plans[].templates[].roles.{role}.capacity_unit_per_instance	Number	Number of capacity units per instance, i.e., base_capacity. When creating a deployment, `capacity` must be an integer multiple of this value.
page_no	Number	Query page number.
page_size	Number	Query page size.
total	Long	Total number of models matching the query conditions.

List deployed models

Get the list of deployed instances, i.e., the model services that are actually running or have been deployed in the current workspace.

Endpoint

GET https://dashscope.aliyuncs.com/api/v1/deployments

Request example

Run the following command to obtain the list of dedicated services:

curl "https://dashscope.aliyuncs.com/api/v1/deployments?page_no=1&page_size=100" \
    --header "Authorization: Bearer ${DASHSCOPE_API_KEY}" \
    --header 'Content-Type: application/json'

Request parameters

Parameter	Type	Location	Required	Description
page_no	Number	query	No	Page number. Default: 1.
page_size	Number	query	No	Page size. Default: 50. Maximum: 200. Minimum: 1.

Response example

Sample response:

{
  "request_id": "7efdd3a7-a90d-96c6-b477-70055d59edf7",
  "output": {
    "page_no": 1,
    "page_size": 10,
    "total": 1,
    "deployments": [
      {
        "deployed_model": "emo-35b3f106-sample01",
        "gmt_create": "2025-06-17T11:00:38",
        "gmt_modified": "2025-06-17T11:06:13",
        "status": "RUNNING",
        "model_name": "emo",
        "base_model": "emo",
        "base_capacity": 1,
        "capacity": 1,
        "ready_capacity": 1,
        "workspace_id": "llm-v71tlv3d***",
        "charge_type": "post_paid",
        "creator": "175805416***",
        "modifier": "175805416***"
      }
    ]
  }
}

Response parameters

Field	Type	Description
request_id	String	Unique identifier for the request.
output.page_no	Number	Current page number.
output.page_size	Number	Number of items per page.
output.total	Number	Total number of deployments.
output.deployments	Array	List of model deployments. Each element has the same fields as the output of Get a model deployment.

Modify deployment throttling

Modify the RPM (Requests Per Minute) and TPM (Tokens Per Minute) throttling settings for a specified model deployment.

Note

Only some models deployed by using model units support modifying RPM and TPM settings.

Endpoint

PUT https://dashscope.aliyuncs.com/api/v1/deployments/{deployed_model}/update

Request example

Run the following command to modify the throttling settings of a specified deployment:

curl -X PUT "https://dashscope.aliyuncs.com/api/v1/deployments/{deployed_model}/update" \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
    "rpm_limit": 1000,
    "tpm_limit": 200
}'

Request parameters

Parameter	Type	Location	Required	Description
deployed_model	String	path	Yes	The unique identifier for the model deployment. You can obtain it from Create deployment or the List deployments operation.
rpm_limit	Number	body	At least one parameter	Requests per minute (RPM).
tpm_limit	Number	body	At least one parameter	Tokens per minute (TPM).

Response example

Sample response:

{
    "request_id": "1d121fd9-876c-40ad-bc40-a9e68ef3b986",
    "output":
    {
        "deployed_model": "qwen-plus-2025-12-01-b6d61c71",
        "gmt_create": "2026-01-07T13:52:44",
        "gmt_modified": "2026-01-07T14:01:41",
        "status": "PENDING",
        "model_name": "qwen-plus-2025-12-01",
        "base_model": "qwen-plus-2025-12-01",
        "base_capacity": 4,
        "capacity": 4,
        "ready_capacity": 0,
        "workspace_id": "llm-8v53e*******",
        "charge_type": "post_paid",
        "creator": "16542902******",
        "modifier": "16542902********",
        "plan": "mu",
        "model_unit_spec": "MU1",
        "enable_thinking": true,
        "max_context_length": 1,
        "rpm_limit": 1000,
        "tpm_limit": 200
    }
}

Response parameters

In addition to the base fields listed in Get a model deployment, the response may also include the following fields:

Field	Type	Description
output.plan	String	Deployment plan.
output.model_unit_spec	String	Model unit deployment template, such as MU1.
output.enable_thinking	Boolean	Whether thinking mode is enabled.
output.max_context_length	Number	Maximum context length.
output.rpm_limit	Number	Requests per minute limit.
output.tpm_limit	Number	Tokens per minute limit.

Scale a deployment

Adjust the number of resource units used by a dedicated service by performing an update operation.

Endpoint

PUT https://dashscope.aliyuncs.com/api/v1/deployments/{deployed_model}/scale

Request example

Run the following command to scale a specified service:

curl --request PUT "https://dashscope.aliyuncs.com/api/v1/deployments/emo-35b3f106-sample01/scale" \
    --header "Authorization: Bearer ${DASHSCOPE_API_KEY}" \
    --header 'Content-Type: application/json' \
    --data '{
                "capacity":2
            }'

Request parameters

Parameter	Type	Location	Required	Description
deployed_model	String	path	Yes	The unique identifier for the model deployment. You can obtain it from Create deployment or the List deployments operation.
capacity	Number	body	Conditionally required	Only available when `"plan": "mu"`. For more information, see feature support for model unit deployment.	The resource units used by the model after the update. Must be an integer multiple of `base_capacity`.
ptu_capacity	Object	body	Conditionally required	Only available when `"plan": "ptu"`. For more information, see feature support for PTU deployment.	Takes effect only when `"plan": "ptu"` is set. Example: `"ptu_capacity": { "input_tpm": 10000, "output_tpm": 1000 }`.
ptu_capacity.input_tpm	Number	body			Supported by all models. Input tokens per minute. The maximum input token amount per minute supported by the deployed model.
ptu_capacity.output_tpm	Number	body			Supported by all models. Output tokens per minute. The maximum output token amount per minute supported by the deployed model.
ptu_capacity.thinking_output_tpm	Number	body			Supported by some models. Thinking output tokens per minute. The maximum thinking output token amount per minute supported by the deployed model.

Response example

Sample response:

{
  "request_id": "6c6b7676-3fea-423b-bc26-c9e2337e1142",
  "output": {
    "deployed_model": "emo-35b3f106-sample01",
    "gmt_create": "2025-06-17T11:00:38",
    "gmt_modified": "2025-06-17T11:42:02.311",
    "status": "UPDATING",
    "model_name": "emo",
    "base_model": "emo",
    "base_capacity": 1,
    "capacity": 2,
    "ready_capacity": 1,
    "workspace_id": "llm-v71tlv3dezezp2en",
    "charge_type": "post_paid",
    "creator": "17580541***",
    "modifier": "17580541***"
  }
}

Response parameters

Response parameters are the same as Get a model deployment. For details, see the response parameters of Get a model deployment.

Delete a deployment

Delete a specified model deployment and release the corresponding computing resources.

Important

After this operation is performed, the model deployment service will be taken offline immediately and cannot be recovered:

The model will no longer be available for invocation.
Billing for the deployment service will stop.

Endpoint

DELETE https://dashscope.aliyuncs.com/api/v1/deployments/{deployed_model}

Request example

Run the following command to delete a specified deployment.

curl --request DELETE "https://dashscope.aliyuncs.com/api/v1/deployments/emo-35b3f106-sample01" \
    --header "Authorization: Bearer ${DASHSCOPE_API_KEY}" \
    --header 'Content-Type: application/json'

Request parameters

Parameter	Type	Location	Required	Description
deployed_model	String	path	Yes	The unique identifier for the model deployment. You can obtain it from Create deployment or the List deployments operation.

Response example

Pay attention to the status field. When the status becomes DELETING, it indicates that the deployment is being deleted.

{
  "request_id": "5378b78b-8564-481f-a3e0-580e551df22c",
  "output": {
    "deployed_model": "emo-35b3f106-sample01",
    "gmt_create": "2025-06-17T11:00:38",
    "gmt_modified": "2025-06-17T11:42:02",
    "status": "DELETING",
    "model_name": "emo",
    "base_model": "emo",
    "base_capacity": 1,
    "capacity": 2,
    "ready_capacity": 1,
    "workspace_id": "llm-v71tlv3***",
    "charge_type": "post_paid",
    "creator": "175805416***",
    "modifier": "175805416***"
  }
}

Then, call Get a model deployment to verify the deletion. If the following response is returned, it means the deployed service no longer exists and has been successfully deleted.

{
    "request_id": "eb619064-0c4f-4d29-aa49-xxxxxx",
    "message": "Not found.",
    "code": "NotFound"
}

Response parameters

Response parameters are the same as Get a model deployment. For details, see the response parameters of Get a model deployment.

Error responses

Response example

{
    "request_id": "ca218d57-b91b-46b2-bd35-c41c6287bcf4",
    "message": "Model: qwen-plus-20230703-cx7f not found!",
    "code": "NotFound"
}

Response parameters

Field	Type	Description
request_id	String	The unique ID for the request.
code	String	The error code.
message	String	The error message.

Possible errors:

Error code	Error message	Error reason
NotFound	Model: xxx not found!	The model specified for the new deployment does not exist. The model associated with the specified deployment does not exist.
Conflict	Deployed model xxx already exists, please specify a suffix.	The specified suffix is already in use.
InvalidParameter	Invalid capacity (xx), capacity must be larger than or equal to 0 and multiples of 1 and less than 1000!	Invalid capacity units specified when creating or updating the deployment.