Overview
Introduction to DataEyesAI Open Platform
Search/Reading Products
Webpage Parsing API
Online Search API and extended applications for weather, stock, events, etc.
Document OCR Parsing API
AI Large Model Products
qwen-image-edit-max
doubao-seedream-5-0-260128
wan2.6-t2i
Video：
doubao-seedance-2-0-260128
wan2.6-i2v-flash
Text：
deepseek-v3.2-251201
glm-5
Multimodal：
z-image-turbo
wan2.6-image
Search/Reading Product Introduction
AI Webpage Parsing and Reading
An AI extraction tool for webpage data that can "read" the visual layout of a webpage like a human, intelligently extracting core content summaries and high-quality corpora, moving away from messy HTML tags.
2.High-Speed Response - Features industry-leading response speed, with average processing time controlled to under 1 second (industry average exceeds 4 seconds), significantly improving data processing efficiency.
3.Leading Extraction Capability - Supports over 15 mainstream webpage types for intelligent extraction, including various dynamic webpages (competitors average support 8-9 types, most unable to handle dynamic content), meeting diverse data acquisition needs.
Enables large models to have an "all-network perspective" instantly. Based on a self-developed search engine system, it can real-time and accurately capture and structure online information while providing visual displays of some data, enhancing human-computer interaction in searches.
1.High Real-Time Performance - Specifically optimized for the Chinese search environment, extracting the latest information.
2.High Relevance - Unlike traditional searches that "search without reading," we deeply analyze search intent, read webpages, and score relevance, providing the most relevant search webpage information.
3.Visual Presentation - Compared to traditional text-only searches, in certain scenarios (such as weather, stocks, events, etc.), we provide real-time information modal cards to enhance display effects.
AI Document OCR Parsing
Specializes in solving various complex PDF and document parsing challenges. Our self-developed model can accurately extract text, tables, and other content, bringing knowledge hidden in documents back to light.
2.High-Precision Extraction - Self-developed model accurately identifies text, tables, and other content types.
3.Knowledge Activation - Transforms dormant data in documents into usable knowledge assets.
1. Text-to-Image Model - qwen-image gemini-3-pro-image
2. Text-to-Video Model - wan 2.6
3. Multimodal Model - claude-opus gemini-3-pro
4. And other model cloud services.
Usage Introduction
HTTP Protocol
Global Request Parameters for Search and Reading
| PARAM | VALUE |
|---|---|
| base_url * | https://api.dataeyes.ai |
| Authorization | Obtainable after registering an account in the console, used for permission verification |
| All global interfaces support HTTP and HTTPS, and each interface requires passing Authorization for permission verification. |
Global Request Parameters for AI Large Models
| PARAM | VALUE | Description |
|---|---|---|
| base_url 1 * | http://cloud.dataeyes.ai | Direct connection line from the source (main site): forwarding recommended, zero latency in Silicon Valley, high-speed and stable |
| base_url 2 * | https://platform.dataeyes.ai | High-speed high-defense dedicated line (international recommendation): CF cluster, high-speed, high-defense, stable, international acceleration |
| base_url 3 * | https://cloud-cn.shuyanai.com | China-US dedicated line: China-US large model dedicated line CDN, suitable for users in mainland China |
| Authorization | Obtainable after registering an account in the console, used for permission verification |
Global Response Structure
| PARAM | VALUE |
|---|---|
| code | Status code |
| data | Main returned data |
| msg | Request status |
Application for Account and Keys
After registration, please enter the console to obtain Authorization, which can be linked to business personnel for communication and inquiries.
Step 2: Complete real-name authentication
In the console - system management module - settings page, based on your type, you can choose personal authentication or corporate authentication. The service call volume provided varies for individuals and enterprises. If you need higher frequency service calls, please contact business to upgrade your account.
Step 3: Account Recharge
Search/Reading Products:
HTTP Request Rate Limiting
Free Version: Suitable for experience and testing
Total call count: 500 times, rate limit: 90 times/min, 3 times/sec;
Starter Version: Suitable for individuals or small enterprises
Total call count: 36,000 times, rate limit: 450 times/min, 15 times/sec;
Professional Version: Suitable for medium enterprises or organizations
Total call count: 1,200,000 times, rate limit: 1,500 times/min, 50 times/sec;
Enterprise Version: Suitable for large enterprises or organizations
Total call count: 6,000,000 times, rate limit: 3,000 times/min, 100 times/sec;
If higher frequency requirements exist, please contact business to upgrade to a higher level of service.