qwen-vl-ocr

Updated at:

Qwen-VL_OCR is an OCR model trained based on Qwen-VL. It aggregates various image-text recognition, parsing, and processing tasks through a unified model approach, offering powerful image-text recognition capabilities.This model version is functionally equivalent to the snapshot model qwen-vl-ocr-2025-11-20.

Inference Service Provider

The inference service provider for qwen-vl-ocr is Alibaba Cloud Model Studio.

Model Capabilities

China (Beijing)

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Supported

Fine-tuning

Unsupported

Singapore

Scope: International

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Germany (Frankfurt)

Scope: Global

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

US (Virginia)

Scope: Global

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter Value Parameter Value

Max Input Length

30000

Max Output Length

8192

Context Window

38192

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

Input(Batch File)

0.15

Per 1M tokens

Output(Batch File)

0.25

Per 1M tokens

Singapore

Scope: International

Billing Item Price (CNY) Unit

Input

0.514

Per 1M tokens

Output

1.174

Per 1M tokens

Germany (Frankfurt)

Scope: Global

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

US (Virginia)

Scope: Global

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

Rate Limits

China (Beijing)

Parameter Value

RPM (Requests Per Minute)

600

TPM (Tokens Per Minute)

6,000,000

Singapore

Scope: International

Parameter Value

RPM (Requests Per Minute)

600

TPM (Tokens Per Minute)

6,000,000

Germany (Frankfurt)

Scope: Global

Parameter Value

RPM (Requests Per Minute)

600

TPM (Tokens Per Minute)

6,000,000

US (Virginia)

Scope: Global

Parameter Value

RPM (Requests Per Minute)

60

TPM (Tokens Per Minute)

100,000

Dynamic Updates

qwen-vl-ocr-latest

Qwen-VL-OCR (qwen-vl-ocr), an OCR recognition large model trained on Qwen-VL. Aggregates multiple image-text recognition, parsing, and processing tasks through a unified model framework, providing powerful image-text recognition capabilities.

Inference Service Provider

The inference service provider for qwen-vl-ocr-latest is Alibaba Cloud Model Studio.

Model Capabilities

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Supported

Fine-tuning

Unsupported

Context Limits

Parameter Value Parameter Value

Max Input Length

30000

Max Output Length

8192

Context Window

38192

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

Input(Batch File)

0.15

Per 1M tokens

Output(Batch File)

0.25

Per 1M tokens

Rate Limits

China (Beijing)

Parameter Value

RPM (Requests Per Minute)

6000

TPM (Tokens Per Minute)

30,000,000

Snapshot Versions

qwen-vl-ocr-2025-11-20

This model is a snapshot version from November 20, 2025, and is based on the latest Qwen-VL3 architecture with a comprehensive upgrade. It features significant improvements in document parsing and text localization capabilities, as well as substantial reductions in end-to-end latency and illusions.

Inference Service Provider

The inference service provider for qwen-vl-ocr-2025-11-20 is Alibaba Cloud Model Studio.

Model Capabilities

China (Beijing)

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Singapore

Scope: International

Capability Support Capability Support

Input Modality

Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Germany (Frankfurt)

Scope: Global

Capability Support Capability Support

Input Modality

Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

US (Virginia)

Scope: Global

Capability Support Capability Support

Input Modality

Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter Value Parameter Value

Max Input Length

30000

Max Output Length

8192

Context Window

38192

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

Singapore

Scope: International

Billing Item Price (CNY) Unit

Input

0.514

Per 1M tokens

Output

1.174

Per 1M tokens

Germany (Frankfurt)

Scope: Global

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

US (Virginia)

Scope: Global

Billing Item Price (CNY) Unit

Input

0.3

Per 1M tokens

Output

0.5

Per 1M tokens

Rate Limits

China (Beijing)

Parameter Value

RPM (Requests Per Minute)

6000

TPM (Tokens Per Minute)

30,000,000

Singapore

Scope: International

Parameter Value

RPM (Requests Per Minute)

1200

TPM (Tokens Per Minute)

6,000,000

Germany (Frankfurt)

Scope: Global

Parameter Value

RPM (Requests Per Minute)

1200

TPM (Tokens Per Minute)

6,000,000

US (Virginia)

Scope: Global

Parameter Value

RPM (Requests Per Minute)

60

TPM (Tokens Per Minute)

100,000

qwen-vl-ocr-2025-08-28

This snapshot version (August 28, 2025) achieves comprehensive upgrades in text localization capability, with enhanced general text recognition and document/invoice information extraction performance.

Inference Service Provider

The inference service provider for qwen-vl-ocr-2025-08-28 is Alibaba Cloud Model Studio.

Model Capabilities

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter Value Parameter Value

Max Input Length

30000

Max Output Length

4096

Context Window

34096

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item Price (CNY) Unit

Input

5

Per 1M tokens

Output

5

Per 1M tokens

Rate Limits

China (Beijing)

Parameter Value

RPM (Requests Per Minute)

600

TPM (Tokens Per Minute)

6,000,000

qwen-vl-ocr-2025-04-13

Qwen-VL-OCR (qwen-vl-ocr-2025-04-13), an OCR recognition large model trained on Qwen-VL. Aggregates multiple image-text recognition, parsing, and processing tasks through a unified model framework, providing powerful image-text recognition capabilities. Built-in six recognition tasks: general recognition, document parsing, table parsing, information extraction, multi-language recognition, and formula recognition. Snapshot version from April 13, 2025.

Inference Service Provider

The inference service provider for qwen-vl-ocr-2025-04-13 is Alibaba Cloud Model Studio.

Model Capabilities

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter Value Parameter Value

Max Input Length

30000

Max Output Length

4096

Context Window

34096

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item Price (CNY) Unit

Input

5

Per 1M tokens

Output

5

Per 1M tokens

Rate Limits

China (Beijing)

Parameter Value

RPM (Requests Per Minute)

600

TPM (Tokens Per Minute)

6,000,000

qwen-vl-ocr-1028

Qwen-VL-OCR (qwen-vl-ocr), an OCR recognition large model trained on Qwen-VL. Aggregates multiple image-text recognition, parsing, and processing tasks through a unified model framework, providing powerful image-text recognition capabilities.

Inference Service Provider

The inference service provider for qwen-vl-ocr-1028 is Alibaba Cloud Model Studio.

Model Capabilities

Capability Support Capability Support

Input Modality

Text Image

Output Modality

Text

Model Experience

Supported

Function Calling

Unsupported

Structured Outputs

Unsupported

Web Search

Unsupported

Prefix Completion

Unsupported

Context Caching

Unsupported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter Value Parameter Value

Max Input Length

30000

Max Output Length

4096

Context Window

34096

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item Price (CNY) Unit

Input

5

Per 1M tokens

Output

5

Per 1M tokens

Rate Limits

China (Beijing)

Parameter Value

RPM (Requests Per Minute)

600

TPM (Tokens Per Minute)

6,000,000