Introduction to
HiDream.ai
CONTENTS
Company Profile
Ecosystem Positioning
Core Technologies
Product lists
Pixeling for Designers
PixMaker for Marketing
HD Studio &
api platform
commercial analysis
global influence
future roadmap
Contact
Company profile
"Responsible intelligence for a better creative future"
Headquarters
Based in Beijing, China
Founded
March 2023
Core Industry
Multi-modal AIGC (Artificial Intelligence Generated Content)
Mission
To empower global creators through responsible, intelligent content-generation tools
Team Composition
Over 90% technical professionals with M.S./Ph.D. degrees
Includes artists, animators, and visual AI experts
Technology Foundation
Proprietary base model “
Pixeling
”
Supports: text-to-image, text-to-video, image-to-video, and entry-level 3D generation
Ecosystem Positioning
Vertical Integration Advantage
Offers consumer-facing platforms, developer APIs, and creative studios
Supports full content lifecycle: generation → editing → deployment
Pixeling Model Ecosystem
Built to serve creators, developers, and enterprises
Enables multi-modal workflows across text, image, and video
Market Focus
From UGC creators to enterprise clients in media, retail, and education
Designed for both Chinese and global audiences
Competitive Edge
Compared to peers like Runway, Pika, and Stability AI, HiDream offers:
Scalable enterprise API infrastructure
Strong community creation layer and prompt-sharing portal
Proactive content safety and ethical model alignment
Core Technologies
🌐 Foundation Model: Pixeling
Supports 4 major modalities: text, image, video, 3D.
Boasts 13B+ parameters, ensuring high fidelity optimized for both speed and quality.
Architecture: Sparse-Diffusion-Transformer + Mixture-of-Experts (MoE).
Open-sourced image model (HiDream-I1) available on Hugging Face.
Industry-leading performance on GenEval and HPSv2.1 benchmarks.
Instruction-Following AI
HiDream-A1: A multi-modal agent for creative instruction tasks.
HiDream-E1: An editing model for image-based tasks (inpainting, enhancement).
Product lists
📍 Pixeling for Designers
Precision Visual API Platform
Click here
💡 PixMaker for Marketing
Light-weight UGC Entry Point
Click here
🎨 HD Studio
Professional Creator Interface
Coming soon…
🧰 api platform
Developer Gateway & Tech Infrastructure
Click here
Pixeling for Designers
Precision Visual API Platform
Positioning:
A high-precision, controllable visual generation platform designed for professional design needs and enterprise-level creative workflows.
Target Users
Visual Designers
Brand Creative Teams
Design Agencies
Platform Developers
Core Features
text to image
Enter any prompt to create images of your dreams.
Image to Image
Create stunning new images from existing ones.
Image Quality Enhancement
Improve the quality of any image in seconds.
Text to video
Type any prompt and watch it come to life.
Image to video
Take image and transform it into a captivating video.
Video to video
Change the style of any video with the help of AI.
Outpainting
Image extension and scene completion capabilities.
Intelligent Editing
AI-based object removal, green screen replacement, and composition optimization.
High-Resolution Output
Commercial-grade image export at high fidelity.
Use Cases
Automated advertising poster creation
Batch brand visual content production
Video platform thumbnail generation
Illustrative sketch assistance
PixMaker for Marketing
Light-weight UGC Entry Point
Positioning: A lightweight, user-friendly AI image generation tool tailored for e-commerce marketing, focusing on product and model image optimization and background enhancement.
Target Users
E-commerce Sellers
Marketing Teams
Brand Managers
Small Studios and Individual Merchants
Core Features
AI background
Generates professional product images with one click.
No need for physical shooting; fast generation supports quick listings.
a. One-click image generation
b. Supports multi-product combination scenes
c. Maintains brand style consistency
AI Model
Creates realistic model images through AI.
No real models required; supports localization and scene variation.
a. Custom digital model creation
b. Model photo generation
- Realistic backgrounds can be switched freely
- Supports global model selection for regional displays
- Output is detailed and natural
Customized Model
Origin Picture
Transformed Picture 1
Transformed Picture 2
AI try on
Applies clothing to any AI model with automatic adaptation. Supports various outfits and realistic fitting effects.
a. Enables try-on with any model
b. Adapts to body shape and posture
c. Covers tops, bottoms, skirts, and dresses
pose switch
Generates multiple model photos in different poses. Keeps the same model and scene in all outputs.
Origin Picture
Transformed Picture 1
Transformed Picture 2
Transformed Picture 3
Product Video
Converts static images into short product videos.
Shows product details and supports video marketing.
remove background
Removes backgrounds from images automatically. Improves focus and adapts to new visual contexts.
AI remove
Removes unwanted objects in one click. Keeps the product visually centered and clean.
image translate
Translates content into multiple languages. Supports cross-border listing and global sales.
coming soon
Adds size labels automatically
Generates promotional visuals for sales
Use Cases
Rapid product launch image creation
Consistent styling for product and lifestyle images
Generating model visuals without physical photoshoots
Bulk social media marketing content production
HD Studio &
api platform
HD Studio – Professional Creator Interface
Positioning: An all-in-one AI-assisted creative platform for illustrators, video creators, and content producers, integrating text-to-image, layered editing, scene construction, and storyboard generation.
Target Users
Illustrators and Comic Artists
Video Content Creators
Creative Writers
Educational Content Developers
coming soon…
API Platform – Developer Gateway & Tech Infrastructure
Positioning: The core AI technology platform providing unified RESTful API services, enabling seamless integration of HiDream’s AIGC capabilities into enterprise and third-party applications.
Target Users
Enterprise Developers
SaaS Providers
Large Client Technical Teams
Startup AI Product Teams
Core Features
RESTful API Access:
Standard HTTP interfaces with POST/GET methods.
Asynchronous Support:
Task queuing and async calls.
Token-Based Authentication:
Secure, granular access control via tokens.
Quota and Rate Limiting:
API usage management with flexible quotas.
Model Versioning:
Support for HiDream-I1 Full, Fast, and Dev versions.
Comprehensive Documentation:
Swagger/OpenAPI specs, code samples, error codes.
Click here
for detailed information
Use Cases
Building enterprise-scale AIGC content platforms
Integration with existing content management systems
Embedding image/video generation into new product workflows
Rapid prototyping of AI-powered creative applications
commercial analysis
Monetization Model
HiDream utilizes a hybrid revenue structure:
• Freemium model: Free credits for new users, upgradable via paid tiers. • Subscription Plans: Monthly/annual for individuals, teams, and enterprises. • API Monetization: Call-based billing for developers and B2B clients. • Custom Deployments: White-label solutions for partners (e.g., media, retail).
Pricing Tiers – API Packages
Plan
Price(¥)
Credits
Validity
Concurrency
Output Capacity (Approx.)
Discount
Trial Plan
100
2000
1 Month
3 tasks
666 images / 28 videos
None
Plan 1
900
10000
1 Year
3 tasks
3,000 images / 142 videos
10% off
Plan 2
8,000
100000
1 Year
5 tasks
33,000 images / 1,428 videos
20% off
Plan 3
35,000
500000
1 Year
7 tasks
166,000 images / 7,000 videos
30% off
Plan 4
60,000
1000000
1 Year
10 tasks
333,000 images / 14,000 videos
40% off
Custom Compute Plans
Contact for business: business@hidream.ai
Service Type
Monthly Price (RMB)
Image
¥8,000
Video
¥15,000
Feature Catalog
Category
Feature
Version
Resolution Options / Description
Cost (pts)
Image Generation
Text-to-Image
v2L
1024×1024, 1248×832, 1360×768, etc.
3
Image-to-Image
v1
2048×2048, 2048×1152, etc.
3
Video Generation
Text-to-Video
v2
960×960, 1280×720, 720×1280
70
Image-to-Video
v2
960×960, 1280×720, 720×1280
70
Image Processing
Smart Expansion
v1
Longest side 1200 px
1
Image Upscale ×2
v1
Max long side 2048 px
2
Image Upscale ×4
v1
Max long side 4096 px
8
Object Removal
v1
Same resolution as original
1
Background Removal
v1
Same resolution as original
2
Image Translation
v1
Same resolution as original
3
AI Try-On
v1
Auto-fit by aspect ratio
17
Marketing Tools
Product Image Gen
v3
2048×2048, 1600×1600, etc.
4
Model Image Gen
v3
Short side 1024 px
8
Model Photo Set
v3
960×1280
4
Image Understanding
Image-to-Text
v1
Convert image to caption
1
gloabl influence
Internationalization & Open Ecosystem
Offers a full English-language interface to support global users
Open-sources key foundation models under MIT license
Accessible via major AI developer platforms: GitHub, Hugging Face, Diffusers
Maintains an active presence at global AI events such as CVPR and WAIC
Ongoing expansion plan targeting Southeast Asia, Japan, and North America markets
Strategic Partnerships & Industry Recognition
Partnerships: Collaborates with Lenovo, China Media Group, Shanghai Film Group, Ciwen Media, and others across media and tech sectors
Awards: Named a Technology Pioneer by the World Economic Forum (2024); listed among Top 50 Chinese AGI Innovators
Key Collaborations:
Provided AI video-generation tools for the Golden Calf Awards
Click here
Supported advanced product visualization pipelines in partnership with Cambricon
Click here
Future Roadmap
Recent Developments (as of Mid-2025)
Released flagship HiDream-I1 (image generation) and E1 (image editing) models
Launched new-generation HD Studio, enabling advanced video creation workflows
Introduced first/last frame + prompt video synthesis, green screen processing
Expanded API coverage with more endpoints and improved real-time serving capabilities
Future Roadmap & Strategic Goals
Enable real-time 3D generation for immersive visual applications
Implement multi-modal prompt chaining: text → image → video → 3D
Expand and localize the global creator community
Scale model architecture to 20B+ parameters for higher fidelity and flexibility
Launch a mobile-native creative suite for portable content production across devices
Contact
WEBSITE
GITHUB
HUGGING FACE
YOUTUBE