OpenAI compatible - Embedding_Alibaba Cloud Model Studio(Model Studio)-Alibaba Cloud Model Studio(Model Studio)-阿里云帮助中心

Alibaba Cloud Model Studio's embedding models are compatible with the OpenAI API. To migrate your existing OpenAI applications to Model Studio, you only need to adjust the following three parameters:

base_url: Replace with https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1.
api_key: Replace with a Model Studio API key.
model: Replace with a model name from the list below.

ImportantAlibaba Cloud Model Studio has released workspace-specific domains for the China (Beijing), Singapore, and China (Hong Kong) regions. The new dedicated domains deliver superior performance and higher stability for inference requests. We recommend migrating to the new domains:

China (Beijing): from https://dashscope.aliyuncs.com to https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com
Singapore: from https://dashscope-intl.aliyuncs.com to https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com

{WorkspaceId} is your workspace ID, which can be found on the Workspace Details page in the Alibaba Cloud Model Studio console. The existing domain remains fully functional.

Supported models

Model	Embedding dimensions	Max rows	Maxtokensper line	Price (per 1K input tokens)	Supported languages	Free quota(Note)
qwen3.7-text-embedding	2,560, 2,048, 1,536, 1,024 (default), 768, 512, 256	20	128,000	CNY 0.0005 Batch call: CNY 0.00025	Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, Russian, and over 200 other major languages and dialects	1 million tokens each Validity: 90 days after you activate Model Studio
qwen3.7-text-embedding-flash	1,024 (default), 768, 512, 256	20	128,000	CNY 0.000125 Batch call: CNY 0.000063	Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, Russian, and over 200 other major languages and dialects
text-embedding-v4 Part of the Qwen3-Embedding series	2,048, 1,536, 1,024 (default), 768, 512, 256, 128, 64	10	8,192	CNY 0.0005 Batch call: CNY 0.00025	Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, Russian, and over 100 other major languages, plus multiple programming languages
text-embedding-v3	1,024 (default), 768, 512, 256, 128, or 64	10	8,192	CNY 0.0005 Batch call: CNY 0.00025	Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, Russian, and over 50 other major languages
text-embedding-v2	1,536	25	2,048	CNY 0.0007 Batch call: CNY 0.00035	Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, Russian	500,000 tokens each Validity: 90 days after you activate Model Studio
text-embedding-v1	1,536	25	2,048	CNY 0.0007 Batch call: CNY 0.00035	Chinese, English, Spanish, French, Portuguese, Indonesian

Multimodal embedding models, such as qwen3-vl-embedding and the tongyi-embedding-vision series, do not support the OpenAI-compatible API. See Multimodal Embedding.

NoteThe OpenAI-compatible API does not support sparse vector output. To use output_type=sparse with text-embedding-v4, call the DashScope API. If output_type=sparse is sent to the OpenAI-compatible API, it returns HTTP 200 with an empty embedding result.

How to use

Call examples

This section provides examples of string input using Python (OpenAI SDK) and cURL (HTTP). For examples that use other programming languages or input methods, see Get started with text embedding.

OpenAI SDK

To call the service using the OpenAI SDK, you must first install the OpenAI SDK.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If you have not configured the environment variable, replace this value with your API key.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1"  # The base_url of the Model Studio service
)

completion = client.embeddings.create(
    model="text-embedding-v4",
    input='The quality of the clothes is excellent. They are beautiful. It was worth the wait. I like them and will buy from here again.',
    dimensions=1024, # Specify the vector dimension. Only text-embedding-v3 and text-embedding-v4 support this parameter.
    encoding_format="float"
)

print(completion.model_dump_json())

HTTP

Submit an interface call

POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1/embeddings

Call from the command line

curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/embeddings' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
    "model": "text-embedding-v4",
    "input": "The quality of the clothes is excellent. They are beautiful. It was worth the wait. I like them and will buy from here again.",
    "encoding_format": "float"
}'

The following result is returned after you run the code:

Result

{
  "data": [
    {
      "embedding": [
        0.0023064255,
        -0.009327292,
        ....
        -0.0028842222
      ],
      "index": 0,
      "object": "embedding"
    }
  ],
  "model":"text-embedding-v4",
  "object":"list",
  "usage":{"total_tokens":23},
  "id":"f62c2ae7-0906-9758-ab34-47c5764f07e2"
}

Sample error response

If an access request error occurs, the cause is indicated by the code and message parameters in the response.

{
    "error": {
        "message": "Incorrect API key provided. ",
        "type": "invalid_request_error",
        "param": null,
        "code": "invalid_api_key"
    }
}

API reference

Text embedding API reference

Error codes

If the model call fails and returns an error message, see Error codes for resolution.