Qwen-VL-OCR

Qwen3-VL 架构模型，用于从扫描件、表格和收据中提取文本和结构化数据。

Alibabaqwen-vl-ocr报告数据错误

规格

上下文: —
最大输出: 8.2K
输入价: $0.07/ 百万 Token
输出价: $0.16/ 百万 Token
发布日期: —

上下文

上下文: —
最大输入: —
最大输出: 8.2K

价格 · USD / 百万 Token

输入: $0.07
输出: $0.16
缓存读取: —
缓存写入: —

输入输出

输入: 文本图像
输出: 文本

API

接口类型: chat

思考

支持思考: 不支持

源信息

状态: 已下线
发布日期: —
知识截止: —

能力

已确认: streaming

调用方式

alibabamodel = qwen-vl-ocr

chat

POSThttps://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1/chat/completions

JSON

qwen-vl-ocr.json

{
  "id": "qwen-vl-ocr",
  "object": "model",
  "created": null,
  "owned_by": "alibaba",
  "name": "Qwen-VL-OCR",
  "api": {
    "types": [
      "chat"
    ]
  },
  "limits": {
    "context": null,
    "input": null,
    "output": 8192
  },
  "modalities": {
    "input": [
      "text",
      "image"
    ],
    "output": [
      "text"
    ]
  },
  "reasoning": {
    "supported": false,
    "efforts": []
  },
  "pricing": {
    "currency": "USD",
    "unit": "1M_tokens",
    "input": 0.07,
    "output": 0.16,
    "cache_read": null,
    "cache_write": null
  },
  "features": [
    "streaming"
  ],
  "info": {
    "status": "retired",
    "release_date": null,
    "knowledge_cutoff": null,
    "description": "Qwen3-VL-architecture model for extracting text and structured data from scans, tables and receipts.",
    "docs": "https://www.alibabacloud.com/help/en/model-studio/vision-model",
    "verified_at": "2026-10-11"
  }
}

官方来源

Vision models | Alibaba Cloud Model Studiowww.alibabacloud.com2026-10-10 Text generation models | Alibaba Cloud Model Studiowww.alibabacloud.com2026-10-10 Model Studio model pricing | Alibaba Cloudwww.alibabacloud.com2026-10-10 [Model Studio] Notice of Retirement for Selected Legacy Long-tail Models | Alibaba Cloudwww.alibabacloud.com2026-10-11