ModelHub
Pay per Token, OpenAI Compatible
Last verified:
What is ModelHub?
ModelHub is an API platform that provides access to 45 Chinese AI models through a single API key, offering DeepSeek V4, GLM-4, Doubao, Kimi, Qwen 3, and DeepSeek R1 among others. The platform delivers AI quality at the same tier as GPT-5.5 while being 97% cheaper, making it an affordable alternative for developers and businesses seeking high-performance AI capabilities.
Key features include a free API key obtained instantly via email signup without requiring a Chinese phone number or credit card, $5 free credit for new users, and a web chat interface at modelhub-api.com/chat that includes 1,000 free requests. The platform supports OpenAI SDK integration with a simple base_url change to https://api.modelhub-api.com/v1, and offers multiple usage methods including environment variables, direct code integration, and curl commands for testing.
ModelHub is designed for developers, AI engineers, and businesses who want access to cutting-edge Chinese AI models without geographic or payment restrictions. The platform serves over 100 million tokens daily with 99.9% uptime and maintains SSL encryption with GDPR compliance. It is particularly valuable for those working on math tasks (where DeepSeek V4 scores 91, equal to GPT-5.5), natural language processing, computer vision, and forecasting applications.
ModelHub pricing
Pricing model: Freemium
Free tier: $5 free credit, no credit card required, 1,000 free web chat requests. Starter plan: $15/month for 60M tokens (~300k pages) with DeepSeek V4 Flash and 1 API key. Pro plan: $65/month for 280M tokens (~1.4M pages) with all models, 5 API keys, and priority support (use code PHLAUNCH50 for 50% off). Enterprise plan: Custom pricing with unlimited tokens, dedicated infrastructure, and 24/7 support.
ModelHub pros
- 97% cheaper than GPT-5.5
- Same AI tier as GPT-5.5
- 45 Chinese AI models available
- $5 free credit for new users
- No credit card required
- No Chinese phone number needed
- Email signup only
- Free 1,000 web chat requests
- OpenAI SDK compatible
- 99.9% uptime guarantee
- 100M+ tokens served daily
- SSL/TLS encrypted traffic
- GDPR compliant
- PayPal, Visa, Mastercard accepted
- Pridefully built by MIT/HKUST/Tsinghua engineers
ModelHub cons
- Limited to Chinese AI models
- No native chat application (requires LobeChat download)
- Starter plan only includes 1 key
- Pro plan at $65/month may be expensive for individuals
- DeepSeek V4 Flash has lowest price but may have limitations
- Enterprise plan pricing is custom/unspecified
- Route through Paddle payment processor (not direct)
- No mention of mobile app availability
Frequently asked questions about ModelHub
Is the AI quality as good as GPT?
DeepSeek V4 Flash scores competitively with GPT-5.5 and Claude Sonnet 4 on independent benchmarks. It's the same tier at a fraction of the cost. DeepSeek V4 math score equals GPT-5.5 (91) and trails by just 4-6 points in other categories.
Do I need a Chinese phone number?
No. Email signup only is required. Global payment is available via Visa/Mastercard/PayPal. There are no Chinese requirements at all for using ModelHub.
Can I use this without downloading anything?
Yes! Just use Terminal (Mac) or Command Prompt (Windows). Copy the curl command from the Developer tab and paste it. The AI responds right there in your terminal without any downloads.
How do you protect my data and payments?
All traffic is SSL/TLS encrypted. Payments are processed by Paddle, a globally trusted payment processor that is PCI Level 1 certified. ModelHub is GDPR compliant and never stores your prompts or uses them for training.
Who runs ModelHub?
ModelHub is run by a dedicated team of engineers and AI researchers with backgrounds from MIT, HKUST, Tsinghua, and USTC. It was founded by Xiao Zhe, an ex-infrastructure engineer, building the bridge between world-class Chinese AI models and global developers.
What models are available?
ModelHub offers 45 Chinese AI models including DeepSeek V4 Flash ($0.15/M tokens), DeepSeek R1 ($0.55/M), Qwen 3 ($0.10/M), GLM-4 ($0.20/M), Doubao ($0.12/M), Kimi ($0.18/M), and 39 more models across NLP, computer vision, and forecasting.
How do I integrate ModelHub with my code?
Replace api_key in your OpenAI SDK with your ModelHub key and change base_url to https://api.modelhub-api.com/v1. You can use environment variables (OPENAI_API_KEY and OPENAI_BASE_URL), direct code integration with the OpenAI client, or test via curl commands.
What is the uptime guarantee?
ModelHub maintains 99.9% uptime with all systems operational. The platform serves over 100 million tokens daily and has 45 models online consistently.
Can I use PayPal for payment?
Yes, ModelHub accepts global payment via Visa, Mastercard, and PayPal. Payments are processed securely through Paddle, which is PCI Level 1 certified.
How many API keys do I get with each plan?
The Starter plan includes 1 API key. The Pro plan includes 5 API keys. Enterprise plans offer dedicated infrastructure with custom key configurations based on your needs.