This topic introduces the features and billing of Image Moderation 2.0.
Introduction to Image Moderation 2.0
Features
The Image Moderation 2.0 API identifies whether images contain content or elements that violate regulations related to internet content distribution, affect platform content order, or impact user experience. It supports 40+ content risk labels and 40+ risk control items. Through Content Moderation's Image Moderation 2.0, you can establish further review or governance measures for specific image content based on industry scenario regulations or platform content governance rules, using the rich risk labels and confidence scores returned by the API.
Version comparison
Image Moderation 2.0 provides more risk types and richer risk labels compared to Image Moderation 1.0, and supports more comprehensive and flexible console configuration features.
Comparison item | Image Moderation 2.0 | Image Moderation 1.0 |
Default QPS | 100 | 50 |
Supported risk detection scope |
| Pornography, politically sensitive and terrorist content, text violations, inappropriate scenes, special logos, QR codes |
Supported risk detection labels | 100+ Note For detailed return labels, see Risk label interpretation table. | 40+ Note For detailed return labels, see label. |
Console |
Note For more feature introductions, see Console operation guide |
|
Supported image formats | PNG, JPG, JPEG, BMP, WEBP, TIFF, SVG, GIF, ICO, HEIC | PNG, JPG, JPEG, BMP, GIF, WEBP |
Billing | Billed by service, with each service detecting multiple risks simultaneously. Cost = Number of images × Number of services Billed separately for the following services:
| Billed by risk scenario (scene). Cost = Number of images × Number of risk scenarios Billed separately for the following risk scenarios:
|
Service description
All services supported by Image Moderation 2.0 are as follows:
Scenario | Service | Detection content | Scenarios |
Common scenario | Baseline Check (baselineCheck_global) | Supports detection of pornography, sexy content, terrorist content, prohibited content, flags, inappropriate content, abuse, special elements, and other risk content in images, including both image content and text content in images (supports Chinese, English, French, German, Indonesian, Malay, Portuguese, Spanish, Thai, Vietnamese, Japanese, Arabic, Filipino, Hindi, Turkish, Russian, Italian, Dutch). For detailed detectable items, see the Content Moderation Console. | Detects whether images contain illegal or inappropriate content for distribution. It is recommended to perform this detection on all images that involve open internet access. |
Moderation large model | Image Moderation for Large and Small Model Integration (postImageCheckByVL_global) | Comprehensively applies image moderation large models and expert model capabilities to comprehensively identify pornography, sexy content, politically sensitive content, terrorist content, prohibited content, religious content, flags, advertising traffic diversion, inappropriate content, abuse, and other violations in images, and can return detailed labels. For detailed detectable items, see the Content Moderation Console. | For image moderation scenarios where the best effect is the priority. It is recommended to choose this service when high effectiveness needs to be guaranteed. For more information, see Large model-based image moderation enhanced service. |
AIGC scenario | AIGC image detection (aigcDetector_global) | Detects the input image during the request to determine whether the image is suspected to be generated by AIGC. | For various scenarios, determines whether images are generated by AIGC. It is recommended to use this when you need to identify the source of images. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service. |
Billing instructions
Pay-as-you-go
When you activate the Image Moderation 2.0 service, the default billing method is pay-as-you-go. You are not charged if you do not call the service. The billing for the Image Moderation 2.0 API is as follows.
Moderation type | Supported business scenarios (services) | Unit price |
Standard image moderation (image_standard) |
| 0.6 USD/thousand calls Note One call to the General baseline detection service is billed once. Billing is based on actual call volume. For example, 100 calls to the General baseline detection service costs 0.06 USD. |
Advanced image moderation (image_advanced) |
| 1.2 USD/thousand calls Note One call to the General baseline detection service is billed once. Billing is based on actual call volume. For example, 100 calls to the Large and small model fusion image moderation service costs 0.12 USD. |
The metering and billing frequency for Content Moderation 2.0 pay-as-you-go is 24 hour(s) per time. In the billing details, moderationType corresponds to the moderation type field above. You can view bill details.
Usage instructions
Connection instructions
Image Moderation Enhanced supports both SDK integration and HTTPS native integration:
For SDK integration, see Image Moderation Enhanced 2.0 SDK and integration guide.
For API native call integration:
For general image detection services, see Image Moderation Enhanced 2.0 synchronous detection API.
For AIGC scenario detection services, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
For image moderation large model services, see Large model-based image moderation service.
Console operation instructions
It is recommended that you modify the image moderation policy in the user console when using it for the first time.
You can modify image moderation detection scope, configure differentiated policies for different businesses, configure custom image libraries and custom dictionaries, view call results, and query usage in the console.
You can perform visual effect testing in the console. It is recommended to perform effect testing after adjusting the image moderation policy.
For specific operations, see Console operation guide.