Referencia de API de Qwen-OCR - Alibaba Cloud Model Studio - Centro de documentación de Alibaba Cloud

Extraiga texto, datos estructurados e información clave de imágenes utilizando el modelo Qwen-OCR. Qwen-OCR admite dos protocolos de API: la API compatible con OpenAI y la API de DashScope.

Para casos de uso y guía de inicio, consulte Extracción de texto (Qwen-OCR).

API compatible con OpenAI

Endpoints

Región	SDK`base_url`	Endpoint HTTP
Singapur	`https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1`	`POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1/chat/completions`
EE. UU. (Virginia)	`https://{WorkspaceId}.us-east-1.maas.aliyuncs.com/compatible-mode/v1`	`POST https://{WorkspaceId}.us-east-1.maas.aliyuncs.com/compatible-mode/v1/chat/completions`
China (Beijing)	`https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1`	`POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/chat/completions`

ImportanteAlibaba Cloud Model Studio ha lanzado dominios específicos de espacio de trabajo para las regiones de China (Beijing) y Singapur. Los nuevos dominios dedicados ofrecen un rendimiento superior y una mayor estabilidad para las solicitudes de inferencia. Recomendamos migrar a los nuevos dominios:

China (Pekín): de https://dashscope.aliyuncs.com a https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com
Singapur: de https://dashscope-intl.aliyuncs.com a https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com

{WorkspaceId} es su ID de espacio de trabajo, que se encuentra en la página Workspace Details de la consola de Alibaba Cloud Model Studio. El dominio existente permanece totalmente funcional.

Requisitos previos

Obtener una clave de API y establecerla como variable de entorno. Si utiliza el SDK de OpenAI, instale el SDK.

Inicio rápido

Utilice el endpoint de completaciones de chat compatible con OpenAI. Envíe un mensaje user con una URL de imagen y un prompt de texto. El modelo extrae el texto y lo devuelve en choices[0].message.content.

Sin transmisión

Python

from openai import OpenAI
import os

PROMPT_TICKET_EXTRACTION = """
Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image.
You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?).
Return the data in JSON format as follows: {'invoice_number': 'xxx', 'departure_station': 'xxx', 'arrival_station': 'xxx', 'departure_date_and_time':'xxx', 'seat_number': 'xxx','ticket_price':'xxx', 'id_card_number': 'xxx', 'passenger_name': 'xxx'},
"""

try:
    client = OpenAI(
        # If the environment variable is not configured, replace with: api_key="sk-xxx"
        api_key=os.getenv("DASHSCOPE_API_KEY"),
        # Singapore region. For US (Virginia), use https://{WorkspaceId}.us-east-1.maas.aliyuncs.com/compatible-mode/v1
        # For China (Beijing), use https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1
        base_url="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
    )
    completion = client.chat.completions.create(
        model="qwen-vl-ocr-2025-11-20",
        messages=[
            {
                "role": "user",
                "content": [
                    {
                        "type": "image_url",
                        "image_url": {"url":"https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg"},
                        # Minimum pixel count. Images below this are upscaled.
                        "min_pixels": 32 * 32 * 3,
                        # Maximum pixel count. Images above this are downscaled.
                        "max_pixels": 32 * 32 * 8192
                    },
                    # Custom prompt. Without this, the model uses: "Please output only the text content from the image without any additional descriptions or formatting."
                    {"type": "text",
                     "text": PROMPT_TICKET_EXTRACTION}
                ]
            }
        ])
    print(completion.choices[0].message.content)
except Exception as e:
    print(f"Error message: {e}")

Node.js

import OpenAI from 'openai';

const PROMPT_TICKET_EXTRACTION = `
Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image.
You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?).
Return the data in JSON format as follows: {'invoice_number': 'xxx', 'departure_station': 'xxx', 'arrival_station': 'xxx', 'departure_date_and_time':'xxx', 'seat_number': 'xxx','ticket_price':'xxx', 'id_card_number': 'xxx', 'passenger_name': 'xxx'}
`;

const client = new OpenAI({
  // If the environment variable is not configured, replace with: apiKey: "sk-xxx"
  apiKey: process.env.DASHSCOPE_API_KEY,
  // For China (Beijing), use https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/compatible-mode/v1
  baseURL: 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1',
});

async function main() {
  const response = await client.chat.completions.create({
    model: 'qwen-vl-ocr-2025-11-20',
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: PROMPT_TICKET_EXTRACTION},
          {
            type: 'image_url',
            image_url: {
              url: 'https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg',
            },
              // Minimum pixel count. Images below this are upscaled.
              "min_pixels": 32 * 32 * 3,
              // Maximum pixel count. Images above this are downscaled.
              "max_pixels": 32 * 32 * 8192
          }
        ]
      }
    ],
  });
  console.log(response.choices[0].message.content)
}

main();

curl

curl -X POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1/chat/completions \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "qwen-vl-ocr-2025-11-20",
  "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "image_url",
                    "image_url": {"url":"https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg"},
                    "min_pixels": 3072,
                    "max_pixels": 8388608
                },
                {"type": "text", "text": "Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image. You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?). Return the data in JSON format as follows: {\'invoice_number\': \'xxx\', \'departure_station\': \'xxx\', \'arrival_station\': \'xxx\', \'departure_date_and_time\':\'xxx\', \'seat_number\': \'xxx\',\'ticket_price\':\'xxx\', \'id_card_number\': \'xxx\', \'passenger_name\': \'xxx\'}"}
            ]
        }
    ]
}'

Transmisión

Establezca stream en true para recibir resultados incrementalmente a medida que el modelo los genera.

Python

import os
from openai import OpenAI

PROMPT_TICKET_EXTRACTION = """
Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image.
You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?).
Return the data in JSON format as follows: {'invoice_number': 'xxx','departure_station': 'xxx', 'arrival_station': 'xxx', 'departure_date_and_time':'xxx', 'seat_number': 'xxx','ticket_price':'xxx', 'id_card_number': 'xxx', 'passenger_name': 'xxx'},
"""

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
)
completion = client.chat.completions.create(
    model="qwen-vl-ocr-2025-11-20",
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image_url",
                    "image_url": {"url":"https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg"},
                    "min_pixels": 32 * 32 * 3,
                    "max_pixels": 32 * 32 * 8192
                },
                {"type": "text","text": PROMPT_TICKET_EXTRACTION}
            ]
        }
    ],
    stream=True,
    stream_options={"include_usage": True}
)

for chunk in completion:
    print(chunk.model_dump_json())

Node.js

import OpenAI from 'openai';

const PROMPT_TICKET_EXTRACTION = `
Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image.
You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?).
Return the data in JSON format as follows: {'invoice_number': 'xxx', 'departure_station': 'xxx', 'arrival_station': 'xxx', 'departure_date_and_time':'xxx', 'seat_number': 'xxx','ticket_price':'xxx', 'id_card_number': 'xxx', 'passenger_name': 'xxx'}
`;

const openai = new OpenAI({
  apiKey: process.env.DASHSCOPE_API_KEY,
  baseURL: 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1',
});

async function main() {
  const response = await openai.chat.completions.create({
    model: 'qwen-vl-ocr-2025-11-20',
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: PROMPT_TICKET_EXTRACTION},
          {
            type: 'image_url',
            image_url: {
              url: 'https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg',
            },
              "min_pixels": 32 * 32 * 3,
              "max_pixels": 32 * 32 * 8192
          }
        ]
      }
    ],
    stream: true,
    stream_options:{"include_usage": true}
  });
  let fullContent = ""
  console.log("Streaming output content:")
  for await (const chunk of response) {
    if (chunk.choices[0] && chunk.choices[0].delta.content != null) {
      fullContent += chunk.choices[0].delta.content;
      console.log(chunk.choices[0].delta.content);
    }
  }
  console.log(`Full output content: ${fullContent}`)
}

main();

curl

curl -X POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1/chat/completions \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "qwen-vl-ocr-2025-11-20",
  "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "image_url",
                    "image_url": {"url":"https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg"},
                    "min_pixels": 3072,
                    "max_pixels": 8388608
                },
                {"type": "text", "text": "Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image. You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?). Return the data in JSON format as follows: {\'invoice_number\': \'xxx\', \'departure_station\': \'xxx\', \'arrival_station\': \'xxx\', \'departure_date_and_time\':\'xxx\', \'seat_number\': \'xxx\',\'ticket_price\':\'xxx\', \'id_card_number\': \'xxx\', \'passenger_name\': \'xxx\'}"}
            ]
        }
    ],
    "stream": true,
    "stream_options": {"include_usage": true}
}'

Parámetros de solicitud

Parámetro	Tipo	Obligatorio	Descripción
`model`	string	Sí	Nombre del modelo. Consulte Modelos recomendados para ver los modelos compatibles.
`messages`	array	Sí	Una matriz de objetos de mensaje que proporciona contexto al modelo.

Objeto de mensaje

Cada mensaje requiere un role (debe ser user) y un arreglo content con estos tipos de elementos:

Parámetro	Tipo	Obligatorio	Descripción
`type`	string	Sí	`text` para entrada de texto, `image_url` para entrada de imagen.
`text`	string	No	El prompt de texto. Valor predeterminado: `"Please output only the text content from the image without any additional descriptions or formatting"`.
`image_url.url`	string	Sí (cuando `type` es `image_url`)	URL o URL de datos codificada en Base64 de la imagen. Para archivos locales, consulte Extracción de texto.
`min_pixels`	integer	No	Umbral mínimo de píxeles. Las imágenes por debajo de este valor se escalan. Consulte Control de resolución de imagen.
`max_pixels`	integer	No	Umbral máximo de píxeles. Las imágenes por encima de este valor se reducen. Consulte Control de resolución de imagen.

Parámetros de generación

Parámetro	Tipo	Predeterminado	Descripción
`stream`	boolean	`false`	Establezca en `true` para recibir respuestas incrementales a medida que el modelo genera la salida.
`stream_options.include_usage`	boolean	`false`	Cuando `stream` es `true`, establezca esto en `true` para incluir el uso de tokens en el último fragmento.
`max_tokens`	integer	Variable	Tokens máximos en la salida. Superar este valor trunca la respuesta. Consulte Límites de tokens de salida.
`temperature`	float	`0.01`	Controla la diversidad de la salida. Los valores más altos producen texto más variado. Rango: [0, 2).
`top_p`	float	`0.001`	Umbral de muestreo de núcleo. Los valores más altos aumentan la diversidad. Rango: (0, 1.0]. Establezca `temperature` o `top_p`, no ambos.
`top_k`	integer	`1`	Limita el conjunto de tokens candidatos durante el muestreo. Si el valor es None o mayor que 100, la política top_k no se habilita y solo entra en vigor la política top_p. Debe ser >= 0. No es un parámetro estándar de OpenAI; páselo mediante `extra_body` en el SDK de Python: `extra_body={"top_k": xxx}`. En el SDK de Node.js o HTTP, páselo en el nivel superior.
`repetition_penalty`	float	`1.0`	Penalización por secuencias repetidas. Los valores superiores a 1.0 reducen la repetición. No es un parámetro estándar de OpenAI; páselo mediante `extra_body` en el SDK de Python.
`presence_penalty`	float	`0.0`	Controla la repetición de contenido. Rango: [-2.0, 2.0]. Los valores positivos reducen la repetición.
`seed`	integer	--	Garantiza resultados reproducibles cuando se utiliza el mismo valor con parámetros idénticos. Rango: [0, 2^31 - 1].
`logprobs`	boolean	`false`	Establézcalo en `true` para devolver las probabilidades logarítmicas de los tokens de salida.
`top_logprobs`	integer	`0`	Número de tokens más probables a devolver por paso. Rango: [0, 5]. Solo efectivo cuando `logprobs` es `true`.
`stop`	cadena o matriz	--	Palabras de parada o ID de tokens. La generación se detiene cuando aparece una cadena especificada o un `token_id`. No mezcle cadenas y `token_id`s en el mismo arreglo.

Respuesta

Respuesta sin transmisión (`chat.completion`)

{
  "id": "chatcmpl-ba21fa91-dcd6-4dad-90cc-6d49c3c39094",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "logprobs": null,
      "message": {
        "content": "```json\n{\n    \"seller_name\": \"null\",\n    \"buyer_name\": \"Cai Yingshi\",\n    \"price_excluding_tax\": \"230769.23\",\n    \"organization_code\": \"null\",\n    \"invoice_code\": \"142011726001\"\n}\n```",
        "refusal": null,
        "role": "assistant",
        "annotations": null,
        "audio": null,
        "function_call": null,
        "tool_calls": null
      }
    }
  ],
  "created": 1763283287,
  "model": "qwen3.5-ocr",
  "object": "chat.completion",
  "service_tier": null,
  "system_fingerprint": null,
  "usage": {
    "completion_tokens": 72,
    "prompt_tokens": 1185,
    "total_tokens": 1257,
    "completion_tokens_details": {
      "accepted_prediction_tokens": null,
      "audio_tokens": null,
      "reasoning_tokens": null,
      "rejected_prediction_tokens": null,
      "text_tokens": 72
    },
    "prompt_tokens_details": {
      "audio_tokens": null,
      "cached_tokens": null,
      "image_tokens": 1001,
      "text_tokens": 184
    }
  }
}

Campo	Tipo	Descripción
`id`	string	Identificador único de solicitud.
`choices`	array	Contenido generado por el modelo.
`choices[].finish_reason`	string	`stop` cuando la generación se completa normalmente, `length` cuando se trunca debido al límite de tokens.
`choices[].index`	integer	Posición en la matriz `choices`.
`choices[].message.content`	string	Texto extraído o salida estructurada del modelo.
`choices[].message.role`	string	Siempre `assistant`.
`choices[].message.refusal`	string	Siempre `null`.
`choices[].message.audio`	object	Siempre `null`.
`choices[].message.function_call`	object	Siempre `null`.
`choices[].message.tool_calls`	array	Siempre `null`.
`created`	integer	Marca de tiempo UNIX de la solicitud.
`model`	string	Modelo utilizado.
`object`	string	Siempre `chat.completion`.
`service_tier`	string	Siempre `null`.
`system_fingerprint`	string	Siempre `null`.
`usage.completion_tokens`	integer	Recuento de tokens de salida.
`usage.prompt_tokens`	integer	Recuento de tokens de entrada.
`usage.total_tokens`	integer	Suma de `prompt_tokens` y `completion_tokens`.
`usage.completion_tokens_details.text_tokens`	integer	Tokens de salida de texto. Otros campos en `completion_tokens_details` son siempre `null`.
`usage.prompt_tokens_details.image_tokens`	integer	Tokens de entrada de imagen.
`usage.prompt_tokens_details.text_tokens`	integer	Tokens de entrada de texto. Otros campos en `prompt_tokens_details` son siempre `null`.

Respuesta con transmisión (`chat.completion.chunk`)

Cuando stream es true, la respuesta se entrega como una serie de fragmentos de Server-Sent Event (SSE). Cada fragmento sigue la misma estructura que la respuesta sin transmisión, con estas diferencias:

object siempre es chat.completion.chunk.
choices[].delta reemplaza a choices[].message. El objeto delta tiene los mismos campos que message.
choices[].delta.role se devuelve solo en el primer fragmento.
finish_reason es null durante la generación, stop al finalizar, o length si se trunca.
Cuando include_usage es true, el último fragmento tiene un arreglo choices vacío e incluye el objeto usage.

{"id":"chatcmpl-f6fbdc0d-78d6-418f-856f-f099c2e4859b","choices":[{"delta":{"content":"","function_call":null,"refusal":null,"role":"assistant","tool_calls":null},"finish_reason":null,"index":0,"logprobs":null}],"created":1764139204,"model":"qwen3.5-ocr","object":"chat.completion.chunk","service_tier":null,"system_fingerprint":null,"usage":null}
{"id":"chatcmpl-f6fbdc0d-78d6-418f-856f-f099c2e4859b","choices":[{"delta":{"content":"```","function_call":null,"refusal":null,"role":null,"tool_calls":null},"finish_reason":null,"index":0,"logprobs":null}],"created":1764139204,"model":"qwen3.5-ocr","object":"chat.completion.chunk","service_tier":null,"system_fingerprint":null,"usage":null}
{"id":"chatcmpl-f6fbdc0d-78d6-418f-856f-f099c2e4859b","choices":[{"delta":{"content":"json","function_call":null,"refusal":null,"role":null,"tool_calls":null},"finish_reason":null,"index":0,"logprobs":null}],"created":1764139204,"model":"qwen3.5-ocr","object":"chat.completion.chunk","service_tier":null,"system_fingerprint":null,"usage":null}
......
{"id":"chatcmpl-f6fbdc0d-78d6-418f-856f-f099c2e4859b","choices":[{"delta":{"content":"","function_call":null,"refusal":null,"role":null,"tool_calls":null},"finish_reason":"stop","index":0,"logprobs":null}],"created":1764139204,"model":"qwen3.5-ocr","object":"chat.completion.chunk","service_tier":null,"system_fingerprint":null,"usage":null}
{"id":"chatcmpl-f6fbdc0d-78d6-418f-856f-f099c2e4859b","choices":[],"created":1764139204,"model":"qwen3.5-ocr","object":"chat.completion.chunk","service_tier":null,"system_fingerprint":null,"usage":{"completion_tokens":141,"prompt_tokens":513,"total_tokens":654,"completion_tokens_details":{"accepted_prediction_tokens":null,"audio_tokens":null,"reasoning_tokens":null,"rejected_prediction_tokens":null,"text_tokens":141},"prompt_tokens_details":{"audio_tokens":null,"cached_tokens":null,"image_tokens":332,"text_tokens":181}}}

Control de resolución de imagen

min_pixels y max_pixels controlan el cambio de tamaño de la imagen antes del procesamiento. La proporción de token a píxel depende de la versión del modelo:

Modelo	Píxeles por token	`min_pixels`predeterminado (mínimo)	`max_pixels`predeterminado	`max_pixels`máximo
`qwen3.5-ocr`, `qwen-vl-ocr-2025-11-20`	32 x 32 = 1,024	3,072 (3 tokens)	8,388,608 (8,192 tokens)	30,720,000 (30,000 tokens)
`qwen-vl-ocr` y anteriores	28 x 28 = 784	3,136 (4 tokens)	6,422,528 (8,192 tokens)	23,520,000 (30,000 tokens)

Comportamiento de redimensionamiento:

Si el recuento de píxeles de la imagen es inferior a min_pixels, la imagen se escala hasta superar min_pixels.
Si el recuento de píxeles de la imagen está dentro de [min_pixels, max_pixels], se utiliza la imagen original sin cambiar su tamaño.
Si el recuento de píxeles de la imagen supera max_pixels, la imagen se reduce por debajo de max_pixels.

Límites de tokens de salida

Modelo	Predeterminado y máximo`max_tokens`
`qwen3.5-ocr`, `qwen-vl-ocr-2025-11-20`, `qwen-vl-ocr-2024-10-28`	Igual a la longitud máxima de salida del modelo. Consulte Selección de modelo.
`qwen-vl-ocr`, `qwen-vl-ocr-2025-11-20`	4,096

Para qwen-vl-ocr, qwen-vl-ocr-2025-11-20, max_tokens tiene como valor predeterminado 4096. Para aumentarlo (4097–8192), contacte a su gerente comercial con: su ID de cuenta de Alibaba Cloud, tipo de imagen (por ejemplo, documentos, comercio electrónico, contratos), nombre del modelo, QPS estimado y volumen diario de solicitudes, y el porcentaje de solicitudes que superan los 4096 tokens de salida.

API de DashScope

Endpoints

Región	Endpoint HTTP
Singapur	`POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation`
EE. UU. (Virginia)	`POST https://{WorkspaceId}.us-east-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation`
China (Beijing)	`POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation`

Configuración de URL base del SDK:

Python:

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

Java (Método 1 -- constructor):

import com.alibaba.dashscope.protocol.Protocol;
MultiModalConversation conv = new MultiModalConversation(Protocol.HTTP.getValue(), "https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1");

Java (Método 2 -- bloque estático):

import com.alibaba.dashscope.utils.Constants;
Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";

Reemplace el dominio con {WorkspaceId}.us-east-1.maas.aliyuncs.com para la región EE. UU. (Virginia) o {WorkspaceId}.cn-beijing.maas.aliyuncs.com para la región China (Beijing). Para la región China (Beijing), no necesita establecer base_url para las llamadas al SDK.

Obtener una clave de API y establecerla como variable de entorno. Si utiliza el SDK de DashScope, también debe instalar el SDK de DashScope.

Tareas integradas

La API de DashScope proporciona tareas de OCR integradas mediante el parámetro ocr_options. Cada tarea utiliza un prompt predeterminado optimizado, lo que elimina la necesidad de un mensaje text.

Tarea	`ocr_options.task`valor	Formato de salida
Reconocimiento general de texto	`text_recognition`	Texto sin formato
Reconocimiento de alta precisión	`advanced_recognition`	Texto sin formato con cuadros delimitadores
Extracción de información	`key_information_extraction`	Pares clave-valor estructurados
Análisis de tablas	`table_parsing`	Estructura de tabla
Análisis de documentos	`document_parsing`	Estructura del documento
Reconocimiento de fórmulas	`formula_recognition`	Fórmulas LaTeX
Reconocimiento multilingüe	`multi_lan`	Texto multilingüe

Reconocimiento de alta precisión

Devuelve texto con datos de posición para cada línea reconocida.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [{
            "role": "user",
            "content": [{
                "image": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241108/ctdzex/biaozhun.jpg",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192,
                "enable_rotate": False}]
            }]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    ocr_options={"task": "advanced_recognition"}
)
print(response["output"]["choices"][0]["message"].content[0]["text"])

Java

// dashscope SDK version >= 2.21.8
import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241108/ctdzex/biaozhun.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.ADVANCED_RECOGNITION)
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("text"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr-2025-11-20",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241108/ctdzex/biaozhun.jpg",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "advanced_recognition"
    }
  }
}
'

Extracción de información

Extrae datos estructurados de clave-valor de imágenes. Especifique los campos a extraer en task_config.result_schema.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [
      {
        "role":"user",
        "content":[
          {
              "image":"http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/demo_ocr/receipt_zh_demo.jpg",
              "min_pixels": 3072,
              "max_pixels": 8388608,
              "enable_rotate": False
          }
        ]
      }
    ]

params = {
  "ocr_options":{
    "task": "key_information_extraction",
    "task_config": {
      "result_schema": {
          "Ride Date": "Corresponds to the ride date and time in the image, in the format YYYY-MM-DD, for example, 2025-03-05",
          "Invoice Code": "Extract the invoice code from the image, usually a combination of numbers or letters",
          "Invoice Number": "Extract the number from the invoice, usually composed of only digits."
      }
    }
  }
}

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    **params)

print(response.output.choices[0].message.content[0]["ocr_result"])

Java

import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;
import com.google.gson.JsonObject;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/demo_ocr/receipt_zh_demo.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        JsonObject resultSchema = new JsonObject();
        resultSchema.addProperty("Ride Date", "Corresponds to the ride date and time in the image, in the format YYYY-MM-DD, for example, 2025-03-05");
        resultSchema.addProperty("Invoice Code", "Extract the invoice code from the image, usually a combination of numbers or letters");
        resultSchema.addProperty("Invoice Number", "Extract the number from the invoice, usually composed of only digits.");

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.KEY_INFORMATION_EXTRACTION)
                .taskConfig(OcrOptions.TaskConfig.builder().resultSchema(resultSchema).build())
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("ocr_result"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr-2025-11-20",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/demo_ocr/receipt_zh_demo.jpg",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "key_information_extraction",
      "task_config": {
        "result_schema": {
          "Ride Date": "Corresponds to the ride date and time in the image, in the format YYYY-MM-DD, for example, 2025-03-05",
          "Invoice Code": "Extract the invoice code from the image, usually a combination of numbers or letters",
          "Invoice Number": "Extract the number from the invoice, usually composed of only digits."
        }
      }
    }
  }
}
'

Análisis de tablas

Extrae la estructura de tablas de las imágenes.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [{
            "role": "user",
            "content": [{
                "image": "http://duguang-llm.oss-cn-hangzhou.aliyuncs.com/llm_data_keeper/data/doc_parsing/tables/photo/eng/17.jpg",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192,
                "enable_rotate": False}]
            }]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    ocr_options={"task": "table_parsing"}
)
print(response["output"]["choices"][0]["message"].content[0]["text"])

Java

import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "https://duguang-llm.oss-cn-hangzhou.aliyuncs.com/llm_data_keeper/data/doc_parsing/tables/photo/eng/17.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.TABLE_PARSING)
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("text"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr-2025-11-20",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "http://duguang-llm.oss-cn-hangzhou.aliyuncs.com/llm_data_keeper/data/doc_parsing/tables/photo/eng/17.jpg",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "table_parsing"
    }
  }
}
'

Análisis de documentos

Extrae el diseño estructural y el texto de los documentos.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [{
            "role": "user",
            "content": [{
                "image": "https://img.alicdn.com/imgextra/i1/O1CN01ukECva1cisjyK6ZDK_!!6000000003635-0-tps-1500-1734.jpg",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192,
                "enable_rotate": False}]
            }]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    ocr_options={"task": "document_parsing"}
)
print(response["output"]["choices"][0]["message"].content[0]["text"])

Java

import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "https://img.alicdn.com/imgextra/i1/O1CN01ukECva1cisjyK6ZDK_!!6000000003635-0-tps-1500-1734.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.DOCUMENT_PARSING)
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("text"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr-2025-11-20",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "https://img.alicdn.com/imgextra/i1/O1CN01ukECva1cisjyK6ZDK_!!6000000003635-0-tps-1500-1734.jpg",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "document_parsing"
    }
  }
}
'

Reconocimiento de fórmulas

Extrae fórmulas matemáticas de imágenes y las devuelve en formato LaTeX.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [{
            "role": "user",
            "content": [{
                "image": "http://duguang-llm.oss-cn-hangzhou.aliyuncs.com/llm_data_keeper/data/formula_handwriting/test/inline_5_4.jpg",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192,
                "enable_rotate": False}]
            }]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    ocr_options={"task": "formula_recognition"}
)
print(response["output"]["choices"][0]["message"].content[0]["text"])

Java

import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "http://duguang-llm.oss-cn-hangzhou.aliyuncs.com/llm_data_keeper/data/formula_handwriting/test/inline_5_4.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.FORMULA_RECOGNITION)
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("text"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "http://duguang-llm.oss-cn-hangzhou.aliyuncs.com/llm_data_keeper/data/formula_handwriting/test/inline_5_4.jpg",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "formula_recognition"
    }
  }
}
'

Reconocimiento general de texto

Extrae texto sin formato de imágenes sin formato estructural.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [{
            "role": "user",
            "content": [{
                "image": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241108/ctdzex/biaozhun.jpg",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192,
                "enable_rotate": False}]
            }]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    ocr_options={"task": "text_recognition"}
)
print(response["output"]["choices"][0]["message"].content[0]["text"])

Java

import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241108/ctdzex/biaozhun.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.TEXT_RECOGNITION)
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("text"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr-2025-11-20",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241108/ctdzex/biaozhun.jpg",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "text_recognition"
    }
  }
}
'

Reconocimiento multilingüe

Reconoce texto en varios idiomas a partir de imágenes.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

messages = [{
            "role": "user",
            "content": [{
                "image": "https://img.alicdn.com/imgextra/i2/O1CN01VvUMNP1yq8YvkSDFY_!!6000000006629-2-tps-6000-3000.png",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192,
                "enable_rotate": False}]
            }]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv('DASHSCOPE_API_KEY'),
    model='qwen-vl-ocr-2025-11-20',
    messages=messages,
    ocr_options={"task": "multi_lan"}
)
print(response["output"]["choices"][0]["message"].content[0]["text"])

Java

import java.util.Arrays;
import java.util.Collections;
import java.util.Map;
import java.util.HashMap;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.aigc.multimodalconversation.OcrOptions;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "https://img.alicdn.com/imgextra/i2/O1CN01VvUMNP1yq8YvkSDFY_!!6000000006629-2-tps-6000-3000.png");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        map.put("enable_rotate", false);

        OcrOptions ocrOptions = OcrOptions.builder()
                .task(OcrOptions.Task.MULTI_LAN)
                .build();
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map
                        )).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .ocrOptions(ocrOptions)
                .build();
        MultiModalConversationResult result = conv.call(param);
        System.out.println(result.getOutput().getChoices().get(0).getMessage().getContent().get(0).get("text"));
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '
{
  "model": "qwen-vl-ocr-2025-11-20",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "image": "https://img.alicdn.com/imgextra/i2/O1CN01VvUMNP1yq8YvkSDFY_!!6000000006629-2-tps-6000-3000.png",
            "min_pixels": 3072,
            "max_pixels": 8388608,
            "enable_rotate": false
          }
        ]
      }
    ]
  },
  "parameters": {
    "ocr_options": {
      "task": "multi_lan"
    }
  }
}
'

Transmisión (DashScope)

Habilite la salida de transmisión para recibir resultados incrementalmente. El método varía según el SDK:

SDK de Python: Establezca stream=True y incremental_output=True.
SDK de Java: Utilice la interfaz streamCall.
HTTP: Establezca el encabezado X-DashScope-SSE: enable.

Python

import os
import dashscope

dashscope.base_http_api_url = 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1'

PROMPT_TICKET_EXTRACTION = """
Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image.
You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?).
Return the data in JSON format as follows: {'invoice_number': 'xxx','departure_station': 'xxx', 'arrival_station': 'xxx', 'departure_date_and_time':'xxx', 'seat_number': 'xxx','ticket_price':'xxx', 'id_card_number': 'xxx', 'passenger_name': 'xxx'},
"""

messages = [
    {
        "role": "user",
        "content": [
            {
                "image": "https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg",
                "min_pixels": 32 * 32 * 3,
                "max_pixels": 32 * 32 * 8192},
            {
                "type": "text",
                "text": PROMPT_TICKET_EXTRACTION
            }
        ]
    }
]

response = dashscope.MultiModalConversation.call(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="qwen-vl-ocr-2025-11-20",
    messages=messages,
    stream=True,
    incremental_output=True,
)
full_content = ""
print("Streaming output content:")
for response in response:
    try:
        print(response["output"]["choices"][0]["message"].content[0]["text"])
        full_content += response["output"]["choices"][0]["message"].content[0]["text"]
    except:
        pass
print(f"Full content: {full_content}")

Java

import java.util.*;

import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversation;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationParam;
import com.alibaba.dashscope.aigc.multimodalconversation.MultiModalConversationResult;
import com.alibaba.dashscope.common.MultiModalMessage;
import com.alibaba.dashscope.common.Role;
import com.alibaba.dashscope.exception.ApiException;
import com.alibaba.dashscope.exception.NoApiKeyException;
import com.alibaba.dashscope.exception.UploadFileException;
import io.reactivex.Flowable;
import com.alibaba.dashscope.utils.Constants;

public class Main {

    static {
        Constants.baseHttpApiUrl="https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1";
    }

    public static void simpleMultiModalConversationCall()
            throws ApiException, NoApiKeyException, UploadFileException {
        MultiModalConversation conv = new MultiModalConversation();
        Map<String, Object> map = new HashMap<>();
        map.put("image", "https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg");
        map.put("max_pixels", 8388608);
        map.put("min_pixels", 3072);
        MultiModalMessage userMessage = MultiModalMessage.builder().role(Role.USER.getValue())
                .content(Arrays.asList(
                        map,
                        Collections.singletonMap("text", "Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image. You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?). Return the data in JSON format as follows: {\'invoice_number\': \'xxx\', \'departure_station\': \'xxx\', \'arrival_station\': \'xxx\', \'departure_date_and_time\':\'xxx\', \'seat_number\': \'xxx\',\'ticket_price\':\'xxx\', \'id_card_number\': \'xxx\', \'passenger_name\': \'xxx\'"))).build();
        MultiModalConversationParam param = MultiModalConversationParam.builder()
                .apiKey(System.getenv("DASHSCOPE_API_KEY"))
                .model("qwen-vl-ocr-2025-11-20")
                .message(userMessage)
                .incrementalOutput(true)
                .build();
        Flowable<MultiModalConversationResult> result = conv.streamCall(param);
        result.blockingForEach(item -> {
            try {
                List<Map<String, Object>> contentList = item.getOutput().getChoices().get(0).getMessage().getContent();
                if (!contentList.isEmpty()){
                    System.out.println(contentList.get(0).get("text"));
                }//
            } catch (Exception e){
                System.exit(0);
            }
        });
    }

    public static void main(String[] args) {
        try {
            simpleMultiModalConversationCall();
        } catch (ApiException | NoApiKeyException | UploadFileException e) {
            System.out.println(e.getMessage());
        }
        System.exit(0);
    }
}

curl

curl --location 'https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--header 'X-DashScope-SSE: enable' \
--data '
{
    "model": "qwen-vl-ocr-2025-11-20",
    "input": {
        "messages": [
            {
              "role": "user",
              "content": [
                  {
                      "image": "https://img.alicdn.com/imgextra/i2/O1CN01ktT8451iQutqReELT_!!6000000004408-0-tps-689-487.jpg",
                      "min_pixels": 3072,
                      "max_pixels": 8388608
                  },
                  {"type": "text", "text": "Please extract the invoice number, train number, departure station, arrival station, departure date and time, seat number, seat class, ticket price, ID card number, and passenger name from the train ticket image. You must accurately extract the key information. Do not omit or fabricate information. Replace any single character that is blurry or obscured by strong light with an English question mark (?). Return the data in JSON format as follows: {\'invoice_number\': \'xxx\', \'departure_station\': \'xxx\', \'arrival_station\': \'xxx\', \'departure_date_and_time\':\'xxx\', \'seat_number\': \'xxx\',\'ticket_price\':\'xxx\', \'id_card_number\': \'xxx\', \'passenger_name\': \'xxx\'}"}
              ]
            }
        ]
    },
    "parameters": {
        "incremental_output": true
    }
}'

Parámetros de solicitud

Parámetro	Tipo	Obligatorio	Descripción
`model`	string	Sí	Nombre del modelo. Consulte Modelos recomendados para ver los modelos compatibles.
`input.messages`	array	Sí	Una matriz de objetos de mensaje.

Objeto de mensaje

Cada mensaje requiere un role (debe ser user) y un campo content (cadena o arreglo). Utilice una cadena para entrada de solo texto. Utilice un arreglo si la entrada incluye datos de imagen, con estos campos:

Parámetro	Tipo	Obligatorio	Descripción
`image`	string	No	URL, URL de datos Base64 o ruta local de la imagen. Consulte Pasar archivos locales.
`text`	string	No	El prompt de texto. Valor predeterminado: `"Please output only the text content from the image without any additional descriptions or formatting"`. No es necesario al utilizar una tarea integrada.
`enable_rotate`	boolean	No	Establézcalo en `true` para corregir imágenes torcidas. Valor predeterminado: `false`.
`min_pixels`	integer	No	Umbral mínimo de píxeles. Consulte Control de resolución de imagen.
`max_pixels`	integer	No	Umbral máximo de píxeles. Consulte Control de resolución de imagen.

Parámetros de generación

Establézcalos en el objeto parameters para llamadas HTTP.

Parámetro	Tipo	Predeterminado	Descripción
`max_tokens`	integer	Variable	Tokens máximos en la salida. Consulte Límites de tokens de salida. En el SDK de Java, use `maxTokens`.
`stream`	boolean	`false`	Habilitar salida con transmisión. Solo SDK de Python. Para Java, use `streamCall`. Para HTTP, establezca `X-DashScope-SSE: enable`.
`incremental_output`	boolean	`false`	Cuando es `true` (recomendado), cada fragmento contiene solo contenido nuevo. Cuando es `false`, cada fragmento contiene la secuencia completa hasta el momento. En el SDK de Java, use `incrementalOutput`.
`temperature`	float	`0.01`	Controla la diversidad de la salida. Rango: [0, 2).
`top_p`	float	`0.001`	Umbral de muestreo de núcleo. Rango: (0, 1.0]. Establezca `temperature` o `top_p`, no ambos.
`top_k`	integer	`1`	Limita el conjunto de tokens candidatos durante el muestreo. Si el valor es None o mayor que 100, la política top_k no se habilita y solo entra en vigor la política top_p. Debe ser >= 0.
`repetition_penalty`	float	`1.0`	Penalización por secuencias repetidas. Los valores superiores a 1.0 reducen la repetición.
`presence_penalty`	float	`0.0`	Controla la repetición de contenido. Rango: [-2.0, 2.0].
`seed`	integer	--	Garantiza resultados reproducibles. Rango: [0, 2^31 - 1].
`logprobs`	boolean	`false`	Establezca en `true` para devolver probabilidades logarítmicas. Modelos compatibles: `qwen-vl-ocr-2025-11-20` y posteriores. En el SDK de Java, utilice el mismo nombre. Para HTTP, colóquelo en `parameters`.
`top_logprobs`	integer	`0`	Número de tokens más probables por paso. Rango: [0, 5]. Solo efectivo cuando `logprobs` es `true`. En el SDK de Java, use `topLogprobs`. Para HTTP, colóquelo en `parameters`.
`stop`	cadena o matriz	--	Palabras de parada o ID de tokens. La generación se detiene cuando aparece una cadena especificada o un `token_id`. No mezcle cadenas y `token_id`s en el mismo arreglo.

Parámetros de tareas integradas (ocr_options)

Al utilizar una tarea integrada, pase ocr_options en parameters (HTTP), como argumento de palabra clave (SDK de Python) o mediante el constructor OcrOptions (SDK de Java).

Parámetro	Tipo	Obligatorio	Descripción
`ocr_options.task`	string	Sí	Nombre de la tarea integrada. Valores válidos: `text_recognition`, `key_information_extraction`, `document_parsing`, `table_parsing`, `formula_recognition`, `multi_lan`, `advanced_recognition`.
`ocr_options.task_config`	object	No	Configuración para `key_information_extraction`.
`ocr_options.task_config.result_schema`	object	No	Objeto JSON que especifica los campos a extraer. Las claves son nombres de campos, los valores son descripciones opcionales para mejorar la precisión. Admite hasta tres niveles de anidamiento.

result_schemaejemplo:

"result_schema": {
     "invoice_number": "The unique identification number of the invoice, usually a combination of numbers and letters.",
     "issue_date": "The date the invoice was issued. Extract it in YYYY-MM-DD format, for example, 2023-10-26.",
     "seller_name": "The full company name of the seller shown on the invoice.",
     "total_amount": "The total amount on the invoice, including tax. Extract the numerical value and keep two decimal places, for example, 123.45."
}

En el SDK de Java, este parámetro es OcrOptions. La versión mínima del SDK de Python de DashScope es 1.22.2. La versión mínima del SDK de Java es 2.18.4. Para advanced_recognition, se requiere SDK de Java >= 2.21.8.

Respuesta

La API de DashScope utiliza un formato de respuesta idéntico para la salida con y sin transmisión.

{
  "status_code": 200,
  "request_id": "8f8c0f6e-6805-4056-bb65-d26d66080a41",
  "code": "",
  "message": "",
  "output": {
    "text": null,
    "finish_reason": null,
    "choices": [
      {
        "finish_reason": "stop",
        "message": {
          "role": "assistant",
          "content": [
            {
              "ocr_result": {
                "kv_result": {
                  "price_excluding_tax": "230769.23",
                  "invoice_code": "142011726001",
                  "organization_code": "null",
                  "buyer_name": "Cai Yingshi",
                  "seller_name": "null"
                }
              },
              "text": "```json\n{\n    \"price_excluding_tax\": \"230769.23\",\n    \"invoice_code\": \"142011726001\",\n    \"organization_code\": \"null\",\n    \"buyer_name\": \"Cai Yingshi\",\n    \"seller_name\": \"null\"\n}\n```"
            }
          ]
        }
      }
    ],
    "audio": null
  },
  "usage": {
    "input_tokens": 926,
    "output_tokens": 72,
    "characters": 0,
    "image_tokens": 754,
    "input_tokens_details": {
      "image_tokens": 754,
      "text_tokens": 172
    },
    "output_tokens_details": {
      "text_tokens": 72
    },
    "total_tokens": 998
  }
}

Campo	Tipo	Descripción
`status_code`	string	`200` indica éxito. El SDK de Java lanza una excepción en lugar de devolver este campo.
`request_id`	string	Identificador único de solicitud. En el SDK de Java, esto es `requestId`.
`code`	string	Código de error. Vacío en caso de éxito. Solo el SDK de Python devuelve este campo.
`output.text`	string	Siempre `null`.
`output.finish_reason`	string	`null` durante la generación, `stop` al completar, `length` cuando se trunca.
`output.choices[].finish_reason`	string	Mismos valores que `output.finish_reason`.
`output.choices[].message.role`	string	Siempre `assistant`.
`output.choices[].message.content[].text`	string	Texto extraído o salida formateada del modelo.
`output.choices[].message.content[].ocr_result`	object	Se devuelve para tareas integradas (`key_information_extraction`, `advanced_recognition`).
`output.choices[].message.content[].ocr_result.kv_result`	object	Resultados de extracción clave-valor (para `key_information_extraction`).
`output.choices[].message.content[].ocr_result.words_info`	array	Resultados de líneas de texto con datos de posición (para `advanced_recognition`).
`output.choices[].message.content[].ocr_result.words_info[].rotate_rect`	array	`[center_x, center_y, width, height, angle]` -- rectángulo delimitador rotado.
`output.choices[].message.content[].ocr_result.words_info[].location`	array	`[x1, y1, x2, y2, x3, y3, x4, y4]` -- cuatro vértices en sentido horario desde la esquina superior izquierda.
`output.choices[].message.content[].ocr_result.words_info[].text`	string	Contenido de la línea de texto.
`output.choices[].message.logprobs`	object	Información de probabilidad logarítmica, devuelta cuando `logprobs` es `true`.
`usage.input_tokens`	integer	Recuento de tokens de entrada.
`usage.output_tokens`	integer	Recuento de tokens de salida.
`usage.characters`	integer	Fijo en 0.
`usage.total_tokens`	integer	Suma de `input_tokens` y `output_tokens`.
`usage.image_tokens`	integer	Tokens correspondientes a la entrada de imagen.
`usage.input_tokens_details.image_tokens`	integer	Tokens de entrada de imagen.
`usage.input_tokens_details.text_tokens`	integer	Tokens de entrada de texto.
`usage.output_tokens_details.text_tokens`	integer	Tokens de salida de texto.

Modelos admitidos

Modelo	Descripción
`qwen3.5-ocr`	Basado en la arquitectura Qwen3.5. Más rápido y preciso. Mejoras importantes en extracción de información, posicionamiento de texto y soporte de conversación de múltiples turnos. Longitud de contexto extendida a 128K.
`qwen-vl-ocr-2025-11-20`	Última instantánea fechada.
	Versión anterior.
`qwen-vl-ocr-2025-11-20`	Versión anterior.
`qwen-vl-ocr-2024-10-28`	Versión anterior.
`qwen-vl-ocr`	Modelo base.

Códigos de error

Si una llamada al modelo devuelve un error, consulte Mensajes de error para resolver el problema.

API compatible con OpenAI

Endpoints

Requisitos previos

Inicio rápido

Sin transmisión

Transmisión

Parámetros de solicitud

Respuesta

Respuesta sin transmisión (chat.completion)

Respuesta con transmisión (chat.completion.chunk)

Control de resolución de imagen

Límites de tokens de salida

API de DashScope

Endpoints

Tareas integradas

Reconocimiento de alta precisión

Extracción de información

Análisis de tablas

Análisis de documentos

Reconocimiento de fórmulas

Reconocimiento general de texto

Reconocimiento multilingüe

Transmisión (DashScope)

Parámetros de solicitud

Respuesta

Modelos admitidos

Códigos de error

Respuesta sin transmisión (`chat.completion`)

Respuesta con transmisión (`chat.completion.chunk`)