Copy

Model overview

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 0.0005Batch 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 eachValidity: 90 days after you activate Model Studio
qwen3.7-text-embedding-flash	1,024 (default), 768, 512, 256	20	128,000	CNY 0.000125Batch 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 0.0005Batch 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 0.0005Batch 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 0.0007Batch call: CNY 0.00035	Chinese, English, Spanish, French, Portuguese, Indonesian, Japanese, Korean, German, Russian	500,000 tokens eachValidity: 90 days after you activate Model Studio
text-embedding-v1	1,536	25	2,048	CNY 0.0007Batch call: CNY 0.00035	Chinese, English, Spanish, French, Portuguese, Indonesian

For model rate limits, see Rate limiting.

Prerequisites

Users familiar with the OpenAI ecosystem can use the OpenAI-compatible API for a quick migration. The DashScope API provides more unique features. Obtain an API key and export the API key as an environment variable. If you use an SDK to make calls, install the DashScope SDK.

OpenAI compatibility

Public cloud

Thebase_urlto configure for SDK calls:https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1Theendpointto configure for HTTP calls:POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/embeddingsReplace {WorkspaceId} with your actual workspace ID.

Request body

modelstringrequiredThe name of the model to call. See the Model overview table for model names.inputarray<string> or string or filerequiredThe input text to process. The input can be a string, an array of strings, or a file. Limits on text length and batch size vary by model version:

qwen3.7-text-embedding model:
- String input: Up to 128,000 tokens.
- String list or file input: Up to 20 items (lines), with each item (line) up to 128,000 tokens.
text-embedding-v3 / v4 models:
- String input: Maximum length of 8,192 tokens.
- String list or file input: A maximum of 10 items (lines), where each item (line) can be up to 8,192 tokens.
text-embedding-v1 / v2 models:
- String input: Maximum length of 2,048 tokens.
- String list or file input: A maximum of 25 items (lines), where each item (line) can be up to 2,048 tokens.

dimensions integer optionalThe dimension of the output embedding vectors. Must be one of the following values: 2560 (for qwen3.7-text-embedding only),2048 (for text-embedding-v4 only), 1536 (for text-embedding-v4 only), 1024, 768, 512, 256, 128, or 64. The default value is 1024.encoding_format string optionalThe returned embedding format. Currently, only float is supported.

Input string
Input string list
Input file

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If the environment variable is not set, replace the placeholder with your API key.
    # China (Beijing) region URL. Replace {WorkspaceId} with your actual workspace ID. URLs differ by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1"
)

completion = client.embeddings.create(
    model="qwen3.7-text-embedding",
    input='The quality of the clothes is excellent and they are beautiful. It was well worth the wait. I love them and will buy from here again.',
    dimensions=1024, # Sets the embedding dimensions (only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4).
    encoding_format="float"
)

print(completion.model_dump_json())

The SDK Expert interactive assistant can accomplish the same development and troubleshooting in natural language. See DashScope SDK Expert.

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If the environment variable is not set, replace the placeholder with your API key.
    # China (Beijing) region URL. Replace {WorkspaceId} with your actual workspace ID. URLs differ by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1"
)

completion = client.embeddings.create(
    model="qwen3.7-text-embedding",
    input=['The wind is strong, the sky is high, and the apes cry mournfully.', 'The islet is clear, the sand is white, and the birds fly back.', 'The boundless forest sheds its leaves shower by shower.', 'The endless river rolls on wave after wave.'],
    dimensions=1024, # Sets the embedding dimensions (only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4).
    encoding_format="float"
)

print(completion.model_dump_json())

Python
Java
curl

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # If the environment variable is not set, replace the placeholder with your API key.
    # China (Beijing) region URL. Replace {WorkspaceId} with your actual workspace ID. URLs differ by region.
    base_url="https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1"
)
# Make sure to replace 'texts_to_embedding.txt' with your file name or path.
with open('texts_to_embedding.txt', 'r', encoding='utf-8') as f:
    completion = client.embeddings.create(
        model="qwen3.7-text-embedding",
        input=f,
        dimensions=1024, # Sets the embedding dimensions (only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4).
        encoding_format="float"
    )
print(completion.model_dump_json())

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.embeddings.CreateEmbeddingResponse;
import com.openai.models.embeddings.EmbeddingCreateParams;

import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.nio.file.Paths;

public class Main {
    public static void main(String[] args) {
        // Create a client using the API key from the environment variable.
        OpenAIClient client = OpenAIOkHttpClient.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                // China (Beijing) region URL. Replace {WorkspaceId} with your actual workspace ID. URLs differ by region.
                .baseUrl("https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1")
                .build();

        // Make sure to replace 'texts_to_embedding.txt' with your file name or absolute path.
        String filePath = "/src/main/java/org/example/text_to_embedding.txt";

        try {
            // Read the file content.
            StringBuilder fileContent = new StringBuilder();
            try (BufferedReader reader = new BufferedReader(new FileReader(filePath))) {
                String line;
                while ((line = reader.readLine()) != null) {
                    fileContent.append(line).append("\n");
                }
            }

            // Create parameters for the vectorization request.
            EmbeddingCreateParams params = EmbeddingCreateParams.builder()
                    .model("qwen3.7-text-embedding")
                    .input(EmbeddingCreateParams.Input.ofString(fileContent.toString()))
                    // Sets the embedding dimensions (only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4).
                    .dimensions(1024)
                    .build();

            // Send the request and get the response.
            CreateEmbeddingResponse response = client.embeddings().create(params);
            System.out.println(response);

        } catch (IOException e) {
            System.err.println("Error reading the file: " + e.getMessage());
            e.printStackTrace();
        } catch (Exception e) {
            System.err.println("Request failed. See the following page for error codes:");
            System.err.println("https://help.aliyun.com/en/model-studio/error-code");
            System.err.println("Error details: " + e.getMessage());
            e.printStackTrace();
        }
    }
}

Replace 'texts_to_embedding.txt' with your file name or path.

FILE_CONTENT=$(cat texts_to_embedding.txt | jq -Rs .)
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/embeddings' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
    "model": "qwen3.7-text-embedding",
    "input": ['"$FILE_CONTENT"'],
    "dimensions": 1024,
    "encoding_format": "float"
}'

Response object

dataarrayA list of the resulting embedding objects.

Property

embeddinglistThe embedding vector, returned as an array of floating-point numbers.indexintegerThe index of the corresponding input text in the input array.objectstringThe object type. The value is always embedding.

modelstringThe name of the model used for this call.objectstringThe object type. The value is always list.usageobjectToken usage. qwen3.7-text-embedding-flash returns prompt_tokens and total_tokens; other existing text embedding models return only total_tokens.

Property

prompt_tokensintegerReturned only by qwen3.7-text-embedding-flash. The total number of tokens in the request input.total_tokens integerThe total number of tokens in the request, used for billing and audit. This count is determined by how the model's tokenizer parses the input string.

idstringA unique request identifier, used for tracing and troubleshooting.

{
  "data": [
    {
      "embedding": [
        -0.0695386752486229, 0.030681096017360687, ...
      ],
      "index": 0,
      "object": "embedding"
    },
    ...
    {
      "embedding": [
        -0.06348952651023865, 0.060446035116910934, ...
      ],
      "index": 5,
      "object": "embedding"
    }
  ],
  "model": "qwen3.7-text-embedding",
  "object": "list",
  "usage": {
    "total_tokens": 184
  },
  "id": "73591b79-d194-9bca-8bb5-xxxxxxxxxxxx"
}

DashScope

Public cloud

base_urlfor SDK calls: https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1Endpoint for HTTP calls: POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/embeddings/text-embedding/text-embeddingReplace {WorkspaceId} with your actual workspace ID.

Request body

modelstringrequiredThe model to use. For a list of available models, see the Model overview table.inputstringorarray<string>requiredThe text to process. The input can be a string, an array of strings, or a file. The supported text length and batch size vary by model version.

qwen3.7-text-embedding and qwen3.7-text-embedding-flash models:
- String input: Up to 128,000 tokens.
- String list or file input: Up to 20 items (lines), with each item (line) up to 128,000 tokens.
text-embedding-v3 / v4 models:
- String input: Up to 8,192 tokens.
- String list or file input: Up to 10 items (lines), with each item up to 8,192 tokens.
text-embedding-v1 / v2 models:
- String input: Up to 2,048 tokens.
- String list or file input: Up to 25 items (lines), with each item up to 2,048 tokens.

text_type stringoptional

When making an HTTP call, place text_typein the parameters object.

Text converted to embeddings can be applied to downstream tasks such as retrieval, clustering, and classification. For asymmetric tasks such as retrieval, it is recommended to differentiate between query text (query) and document text (document) to achieve better retrieval performance. For symmetric tasks such as indexing, clustering, and classification, you can simply use the system default value of document.dimension integer optional

When making an HTTP call, place dimensionin the parameters object.

Specifies the embedding dimension for the output vector. Valid values are 2560 (for qwen3.7-text-embedding only),2048 (for text-embedding-v4 only), 1536 (for text-embedding-v4 only), 1024, 768, 512, 256, 128, or 64. Defaults to 1024.output_type string optional

When making an HTTP call, place output_typein the parameters object.

Specifies the output vector type. This parameter applies only to the qwen3.7-text-embedding,``text-embedding-v3, and text-embedding-v4 models. Valid values are dense, sparse, and dense&sparse. Defaults to dense, which returns only the dense vector representation. Sparse vector outputs (sparse and dense&sparse) are supported only through the DashScope API.instruct string optionalProvides custom instructions to guide the model in understanding the query intent. English instructions are recommended, as they typically improve performance by 1% to 5%.

Input string
Input string list
Input file

Python

import dashscope
from http import HTTPStatus
# China (Beijing) region configuration. Replace {WorkspaceId} with your actual workspace ID. Configurations differ by region.
dashscope.base_http_api_url = "https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1"

resp = dashscope.TextEmbedding.call(
    model="qwen3.7-text-embedding",
    input='The quality of the clothes is excellent and they are beautiful. It was well worth the wait. I love them and will buy from here again.',
    dimension=1024,  # Specifies the output embedding dimension. This parameter is available only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4 models.
    output_type="dense&sparse"
)

print(resp) if resp.status_code == HTTPStatus.OK else print(resp)

Python

import dashscope
from http import HTTPStatus
# China (Beijing) region configuration. Replace {WorkspaceId} with your actual workspace ID. Configurations differ by region.
dashscope.base_http_api_url = "https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1"

DASHSCOPE_MAX_BATCH_SIZE = 10

inputs = ['A swift wind, a high sky, and the gibbons cry mournfully.', 'A clear islet, white sand, and the birds fly back.', 'Boundless rustling woods shed their leaves.', 'The endless Yangtze River comes rolling in.']

result = None
batch_counter = 0
for i in range(0, len(inputs), DASHSCOPE_MAX_BATCH_SIZE):
    batch = inputs[i:i + DASHSCOPE_MAX_BATCH_SIZE]
    resp = dashscope.TextEmbedding.call(
        model="qwen3.7-text-embedding",
        input=batch,
        dimension=1024  # Specifies the output embedding dimension. This parameter is available only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4 models.
    )
    if resp.status_code == HTTPStatus.OK:
        if result is None:
            result = resp
        else:
            for emb in resp.output['embeddings']:
                emb['text_index'] += batch_counter
                result.output['embeddings'].append(emb)
            result.usage['total_tokens'] += resp.usage['total_tokens']
    else:
        print(resp)
    batch_counter += len(batch)

print(result)

Python
Java
curl

from http import HTTPStatus
from dashscope import TextEmbedding
# China (Beijing) region configuration. Replace {WorkspaceId} with your actual workspace ID. Configurations differ by region.
dashscope.base_http_api_url = "https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1"

with open('texts_to_embedding.txt', 'r', encoding='utf-8') as f:
    resp = TextEmbedding.call(
        model="qwen3.7-text-embedding",
        input=f,
        dimension=1024 # Specifies the output embedding dimension. This parameter is available only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4 models.
    )

    if resp.status_code == HTTPStatus.OK:
        print(resp)
    else:
        print(resp)

import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import com.alibaba.dashscope.embeddings.TextEmbedding;
import com.alibaba.dashscope.embeddings.TextEmbeddingParam;
import com.alibaba.dashscope.embeddings.TextEmbeddingResult;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.utils.Constants;

public final class Main {
    public static void main(String[] args) {
        // China (Beijing) region configuration. Replace {WorkspaceId} with your actual workspace ID. Configurations differ by region.
        Constants.baseHttpApiUrl = "https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1";
        try (BufferedReader reader = new BufferedReader(new FileReader("<path_to_your_content_root>"))) {
            StringBuilder content = new StringBuilder();
            String line;
            while ((line = reader.readLine()) != null) {
                content.append(line).append("\n");
            }

            TextEmbeddingParam param = TextEmbeddingParam.builder()
                    .model("qwen3.7-text-embedding")
                    .text(content.toString())
                    .parameter("dimension", 1024)  // Specifies the output embedding dimension. This parameter is available only for qwen3.7-text-embedding, text-embedding-v3, and text-embedding-v4 models.
                    .build();

            TextEmbedding textEmbedding = new TextEmbedding();
            TextEmbeddingResult result = textEmbedding.call(param);

            if (result != null) {
                System.out.println(result);
            } else {
                System.out.println("Failed to get embedding: " + result);
            }
        } catch (IOException | ApiException | NoApiKeyException e) {
            e.printStackTrace();
        }
    }
}

Replace 'texts_to_embedding.txt' with your file name or path.

FILE_CONTENT=$(cat texts_to_embedding.txt | jq -Rs .)
curl --location 'https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/embeddings/text-embedding/text-embedding' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
    "model": "qwen3.7-text-embedding",
    "input": {
        "texts": ['"$FILE_CONTENT"']
    },
    "parameters": {
        "dimension": 1024,
        "output_type": "dense"
    }
}'

Response object

status_code stringThe HTTP status code. A value of 200 indicates success.request_id stringA unique identifier for the request. Use this ID to trace and troubleshoot the request.code stringThe error code returned if the request fails. This field is empty for successful requests.message stringA detailed error message if the request fails. This field is empty for successful requests.outputobjectThe result of the task.

Properties

embeddingsarrayThe model's output for the request. This is an array of objects, with each object corresponding to an input text.

Properties

sparse_embedding arrayThe sparse vector representation of the corresponding string. This applies only to text-embedding-v3 and text-embedding-v4.

Properties

index integerThe index of the token in the vocabulary.value floatIndicates the weight or importance score of the Token. The higher the value, the greater the importance or relevance of the Token in the current text context.token stringThe text of the token.

embedding arrayThe dense vector representation for the corresponding string.text_index integerThe index of the corresponding text in the input array.

usageobjectToken usage. qwen3.7-text-embedding-flash returns prompt_tokens and total_tokens; other existing text embedding models return only total_tokens.

Properties

prompt_tokens integerReturned only by qwen3.7-text-embedding-flash. The total number of tokens in the request input.total_tokens integerThe total number of tokens in the request, used for billing and audit. This count is determined by how the model's tokenizer parses the input string.

{   "status_code": 200,
    "request_id": "1ba94ac8-e058-99bc-9cc1-7fdb37940a46",
    "code": "",
    "message": "",
    "output":{
        "embeddings": [
          {
             "sparse_embedding":[
               {"index":7149,"value":0.829,"token":"swift"},
               .....
               {"index":111290,"value":0.9004,"token":"mournfully"}],
             "embedding": [-0.006929283495992422,-0.005336422007530928, ...],
             "text_index": 0
          },
          {
             "sparse_embedding":[
               {"index":246351,"value":1.0483,"token":"islet"},
               .....
               {"index":2490,"value":0.8579,"token":"back"}],
             "embedding": [-0.006929283495992422,-0.005336422007530928, ...],
             "text_index": 1
          },
          {
             "sparse_embedding":[
               {"index":3759,"value":0.7065,"token":"Boundless"},
               .....
               {"index":1130,"value":0.815,"token":"leaves"}],
             "embedding": [-0.006929283495992422,-0.005336422007530928, ...],
             "text_index": 2
          },
          {
             "sparse_embedding":[
               {"index":562,"value":0.6752,"token":"endless"},
               .....
               {"index":1589,"value":0.7097,"token":"in"}],
             "embedding": [-0.001945948973298072,-0.005336422007530928, ...],
             "text_index": 3
          }
        ]
    },
    "usage":{
        "total_tokens":27
    }
}

Error codes

If a model call fails, see Error Messages.

​Model overview

​Prerequisites

​OpenAI compatibility

Request body

Response object

​DashScope

Request body

Response object

​Error codes

Model overview

Prerequisites

OpenAI compatibility

DashScope

Error codes